Branding and Advertising [1 ed.] 9788763099608, 9788763001182

Citation preview

Branding and Advertising EDITED BY FLEMMING HANSEN & LARS BECH CHRISTENSEN

Branding and Advertising © Copenhagen Business School Press Printed in Denmark 1. edition 2003 e-ISBN 978-87-630-9960-8

Distribution: Scandinavia DBK, Mimersvej 4 DK-4600 Køge, Denmark Phone: +45 3269 7788, fax: +45 3269 7789

North America International Specialized Book Services 920 NE 58th Ave., Suite 300 Portland, OR 97213, USA Tel +1 800 944 6190 Fax +1 503 280 8832 Email: [email protected]

Rest of the World Marston Book Services, P.O. Box 269 Abingdon, Oxfordshire, OX14 4YN, UK Phone: +44 (0) 1235 465500, fax: +44 (0) 1235 465655 E-mail: [email protected]

All rights reserved. No part of this publication may be reproduced or used in any form or by any means - graphic, electronic or mechanical including photocopying, recording, taping or information storage or retrieval systems - without permission in writing from Copenhagen Business School Press at www.cbspress.dk

Table of contents Introduction: Branding and Advertising

8

BRANDING AND COMMUNICATION 1 Advertising and Brand Equity LARRY PERCY The Nature of Brands 12 / Brand Positioning 15 / Summary 21 2 Brands are just like real people! The development of SWOCC’s Brand Personality Scale EDITH G. SMIT, EMILIE VAN DEN BERGE AND GIEP FRANZEN Brand personality in theory 23 / Brand personality measurement 26 / Overview research project 28 / Results 33 / Discussion 38 / Appendix 40 3 Behavioral Finance-based advertising research in the mutual fund industry KLAUS PETER KAAS AND JENNY JORDAN Advertising in the Mutual Fund Industry 44 / Risk-ReturnPerceptions as Variables of Advertising Effects 45 / Relevance of Private Investors’ Judgmental Heuristics for Risk-Return Perceptions and Advertising Content 47 / Experimental Study 50 / Discussion 56 / Appendix 57

12

22

44

4 An evaluation of corporate brand character PERNILLE SCHNOOR Background 59 / Corporate brands and credibility 61 / The study 63 / Conclusion and implications 68

59

5 Barriers to e-Commerce GORM KUNØE Literature Review 72 / The Research Agenda 73 / Method 78 / Results 79 / Discussion 82

70

6 Advertising research: a case study of Lithuanian breweries LAIMONA SLIBURYTE AND REGINA VIRVILAITE Background 85 / Purpose of research 93 / Methodology 94 / Findings 94 / Conclusions 106 / Appendix 109

84

Table of contents 7 Advertising and the Prominence of the Corporate Brand Identifying Brand Architecture Use through Content Analysis KIM CRAMER, PETER C. NEIJENS AND EDITH G. SMIT The Role of the Corporate Brand in Corporate Image 113 / The Role of the Corporate Brand in Brand Portfolio Management 115 / Advertising Budget 117 / Degree of Fit between Products and Core Business 119 / Globalization 120 / Study 120 / Corporate Brand Prominence: Content Analysis 120 / Codebook 121 / Company Information: Desk Research and Interviews 124 / Results 125 / Corporate Brand Prominence in Advertising 127 / Determinants of Corporate Brand Prominence 128 / Other Results 130

111

ADVERTISING EFFECTS 8 Analyzing Effects of Advertising Using Conditional Logistic Regression KRISTINA BIRCH Introduction 133 / Logistic Regression 134 / Conditional Logistic Regression 137 / AdLab Results 140 / Conclusion 146 9 Modelling Purchases as a Function of Advertising and Promotion FLEMMING HANSEN, LOTTE YSSING HANSEN AND LARS GRØNHOLDT Introduction 148 / The data material 149 / The methodology 154 / Presentation of results 158 / Summary 163 / Appendix 165

4

133

148

10 What do Art Directors think the Effects of Advertising are? KJELL GRØNHAUG Introduction 168 / Thinking, Knowledge and Doing 169 / Research Methodology 172 / Findings 174 / Concluding Remarks 176

168

11 Program Context Effects on Commercial Processing MARJOLEIN MOORMAN, PETER C. NEIJENS AND EDITH G. SMIT Program Context Effects on Recall and Attention towards Television Commercials 177 / Psychological Responses and Advertising Processing 178 / Mediation of Variations in Commercial Placement 180 / Method 181 / Results 183 / Discussion 187

177

Table of contents 12 Sports advertising: a review of perimeter advertising effectiveness BJÖRN WALLISER Methodology: Identification of relevant articles and analysis 192 / Image (and attitudinal) effects of perimeter advertising 199 / Discussion 201 / Limitations and future research 203 13 Advertising and the Image of Politicians WOJCIECH CWALINA AND ANDRZEJ FALKOWSKI The Image 206 / Political image formation: A constructivist approach 209 / Political image formation: A realist approach 227 / Practical implications 230

189

205

MEDIA AND TARGETS 14 New Subtle Advertising Formats: Characteristics, Causes & Consequences ROY LANGER Promotional seduction and Fordism – two sides of the same coin 234 / Current trends in advertising strategies, formats and contents 241 / The practitioners’ perspective: professionals on current trends in advertising 250 / Discussion 258 15 Audience Reactions towards Non-Spot Advertising: Influence of Viewer and Program Characteristics PETER C. NEIJENS AND EDITH G. SMIT Introduction 267 / Different forms of non-spot advertising 267 / Background for the growth in non-spot advertising 268 / Public reactions to non-spot advertising: findings from the literature 268 / Research design 273 / Audience research 273 / The nature and extent of non-spot advertising 274 / Audience reactions 275 / Conclusions 282 16 Conceptualising Television Advertising LARS PYNT ANDERSEN Genre 285 / The genre matrix 290 / Genre evolution? 295 / Ironic Selling Propositions 298 / Possible advantages of ISP 303 / Research perspectives 304

232

266

284

5

Table of contents 17 The Role of Lifestyle and Personality in Explaining the Attitude toward Ads and the Purchase Intentions SANDRA DIEHL AND RALF TERLUTTER Introduction 306 / Research on the effects of advertising 307 / Individual differences 309 / Personality and advertising 312 / Lifestyle and advertising 314 / Method 316 / Results 318 / Lifestyle 320 / Discussion 327 / Further research 329 / Summary 330 18 Television advertising and children: A global perspective HANNE NISS Introduction 332 / Children’s media culture and advertising 333 / Global advertising and regulatory environment 335 / Theories of influence 335 / Literature review 336 / Research questions 338 / Methodology 339 / Results 340 / Discussion 341 / Conclusion 344 19 A Changed Picture of Children on Television How and Why? CECILIA VON FEILITZEN Main questions 347 / Samples and method 349 / Findings interpreted in theoretical perspectives 350 / A wider view 355

306

332

346

LOW INVOLVEMENT AND EMOTIONAL RESPONSES 20 Low Involvement Processing How advertising works at low attention levels ROBERT HEATH A summary of the Low Involvement Processing model 356 / Implicit Learning 358 / The limitations of Implicit Memory 360 / Case study 362 / Discussion 363 21 The Nature of Central and Peripheral Advertising Information Processing FLEMMING HANSEN AND LOTTE YSSING HANSEN Low involvement information processing 366 / The ELAM Model 367 / Application of the model 370 / Differences between central and peripheral information processing 376 / Concluding remarks 384

6

356

365

Table of contents 22 A Note on the Role of Advertising in Memory Creation and Memory Reconstruction of Experiences RITA MÅRTENSON Introduction and Background 387 / A Societal View of Experiences 392 / A Commercial View of Experiences 395 / Definition of Experiences in this Chapter 396 / Experiences in a Memory Perspective 398 / The Role of Advertising in Memory Creation and Memory Reconstruction 403 / Summary and Suggestions for Future Research 408

387

References

410

About the Authors

470

7

Introduction:

Branding & Advertising

In 2002, at the Copenhagen Business School, the Center for Marketing Communication organized the First International Conference on Research in Advertising. The purpose of the conference was to create a forum, where people concerned with advertising research in the academic world could meet and exchange views, and where they could meet with practitioners experienced with advertising research in the commercial world. Up until then, advertising researchers had had to rely upon general consumer behaviour, marketing, communication, economic psychology, and similar conferences for presenting their research and having professional feed-back on their work. This rarely created interesting discussion on research topics with colleagues working in similar lines. There may have been many reasons for this state of affairs. One being the low esteem associated with advertising research by other researchers in communication and related areas. Another being the almost unbreakable wall that existed between what went on in commercial studies of media, advertising effects, advertising budgeting, etc. on the one hand, and academicians’ attempts to gain insight to how advertising works, how it should be characterised, and how its role in society should be appreciated. The Copenhagen event and the present publication demonstrates an obvious need for a forum of the kind where people in different locations can get together, and exchange views on important advertising related issues. The material in the present publication represents chapters that have been developed from presentations at the Copenhagen meeting, together with a few other, relevant contributions from researchers in the area. The title of the book, “Branding and Advertising”, was not decided upon prior to the planning of the conference, but as the 8

Introduction: Branding and Advertising material came in and was organized it became evident that this would be the proper denominator for the material presented here. In general, advertising and advertising research may be viewed as covering problems relating to • Determination of the advertising budget • Choice of media group • Development of advertising message • Timing of the campaign. There are contributions in the present book, relating to practically all of these issues, but almost all the contributions are concerned with branding issues, or at least have an undertone of this. Advertising may work in many other contexts than for specific branding purposes. Future conferences on the topic may bring more of those into light. Research on public advertising, and on advertising in social aid organisations, and very specific advertising topics, such as advertising for recruiting and for real estate, may be just a few such examples. The material in the present book is organized in four sections. The first section is titled “Branding and Communication”. Here, the issues on brand equity, brand personality, corporate brand character, and brand architecture, are central topics. Some of these are seen in the light of consumer behaviour brands, others relating to the mutual funds industry, e-commerce, branding in Eastern Europe, and studies of the content of corporate brand advertising. The second major section is concerned with advertising effects. Here, papers are presenting different models applied to single-source data, with the purpose of demonstrating advertising effects, studies of recall and attention in different kinds of contexts, advertising effects, and effects of political advertising. In the third section focus is changed to media and target groups. Here, concern is with new forms of advertising, advertising to children, and structuring the content of particularly television advertising. Finally, in the fourth section an important issue is brought up, which we think will become a major topic in future research. Emotional processes, low-involvement effects, central and peripheral advertising, and different kinds of memory for advertising are looked upon.

9

Introduction: Branding and Advertising Throughout the book, the contributions demonstrate a concern for a number of very varied topics, relating to the functioning of advertising. They also demonstrate the lack of a total, integrated theory in the area. The traditional effect hierarchy thinking is reduced to one out of several possible advertising effect models. The need for workable tools for classifying different kinds of advertising is demonstrated, and the need for seeing how advertising may function very differently, depending upon such things as involvement, fast moving consumer goods versus durables and services, new introductions versus ongoing competitive advertising, developing new media and the changing nature of existing ones, etc. An undertone present in most of the presentations reflects the need for understanding also the role of advertising in a more societal context. In resent years, advertising has undergone dramatic changes. Some talk about the “end of advertising as we know it”. Other concerns themselves with emergence of still more and different media in which advertising appear. We are living in a world where the role of advertising is changing dramatically. The need for a forum where this can come up for debate is evident, and the need for a forum for people to meet who are concerned with these issues is obvious. At the time of writing these introductory remarks to “Branding and Advertising” the Second International Conference on Research in Advertising has already been held in Amsterdam in June 2003. The variety of contributions presented here, the number of researchers attending, and the variety of issues taken up, confirm the relevance of the present, introductory remarks. The book “Branding and Advertising” is a product of contributions from a number of people. The editors only represent the final organizers of the publication. Many thanks should be given to research assistants Jens Halling, Lotte Yssing Hansen, Jeanette Rasmussen, Pernille Christiansen, and Morten Hallum Hansen, for their invaluable contributions in organizing and finalising the programme and the material for the book. Also thanks to our colleagues at the Marketing department at the Copenhagen Business School: Jens Carsten Nielsen, Lars Grønholdt, and many others. Much appreciation should be given also to our colleagues at ASCoR, the Amsterdam School of Communications Research at the University of Amsterdam for their interest in the programme, and for 10

Introduction: Branding and Advertising their willingness to carry on the idea of organizing advertising research conferences. Our thanks goes to all our Dutch colleagues, but a special, warm appreciation should be expressed towards Peter Neijens for his contribution to the present book and his future related work. Finally, we should thank “Dansk Erhvervslivs Pris for Afsætningsøkonomisk Forskning”, “Foreningen til Unge Handelsmænds Uddannelse” and special funds at the Copenhagen Business School, making the organisation of the conference and particularly the publication of this book possible.

Copenhagen, August 2003 Flemming Hansen Professor, ekon.dr.

Lars Bech Christensen Research Assistant, M.Sc.

11

1

Advertising and Brand Equity LARRY PERCY

Advertising and Brand Equity To understand the relationships between advertising and brand equity, it pays to first understand just what we mean by a ‘brand.’ The original meaning of the word ‘brand’ seems to derive from an Old Norse word brandr, which meant ‘to burn’ (Interbrand Group, 1992). Yet in the etymology of the word, this idea of branding as a “permanent mark deliberately made with hot iron” now takes second place to “goods of particular name or trade mark” (Oxford English Dictionary, 1990). But does this really describe what we understand as a brand? The American Marketing Association describes a brand as a “name, term, sign, symbol, or design, or a combination of them intended to identify the goods and services of one seller or group of sellers and differentiate them from those of competitors.” The AMA definition reminds us of the reason for a brand: to enable a person to identify one alternative from a competitor. All of this is true, but a brand must be a label in the true sense of that word: something “attached to an object to give information about it” (Oxford English Dictionary, 1990). But how is that information communicated?

The Nature of Brands When we think of brands, we usually think of products we buy: Coke, Cadbury, Ford, Hoover, Persil, and Mars. But just about anything can

12

Advertising and Brand Equity be ‘branded.’ Products, services, corporations, retail stores, cities, organizations, even individuals can be seen as ‘brands.’ Remember, a brand name is meant to embody information about something, information that represents an added value, differentiating it in a marked way from alternatives. A brand name is meant to trigger in memory positive associations with that brand. Politicians, hospitals, entertainers, football clubs, corporations, all want their name, their brand, to mean something very specific to their market. It is how they wish to be seen, and how they wish to be distinguished from competitive alternatives.

Brand Attitude A brand does provide information. But what kind of information does a brand provide, and where does it come from? Think about some brands you know. What comes to mind when you think about them? No doubt a great deal more than the fact that it is a particular product. Perhaps you were thinking about how much you like it, that it is well known, or that it is ‘one of the best.’ All of these thoughts reflect what we call brand attitude. A brand name represents everything a person knows about a particular product and what it means to them. It provides a convenient summary of their feelings, knowledge and experience with the brand. It means they do not need to spend a great deal of time ‘researching’ a product each time they are considering a purchase. A person’s evaluation of a product is immediately reconstructed from memory, cued by the brand name. But again, where does that brand attitude come from?

Brand Equity The effect of a positive brand attitude leads to something marketers call brand equity. What exactly is brand equity? Most marketers would agree that it is that ‘something’ attached to a brand that adds value over and above the objective characteristics of the product or service. Whatever that ‘something’ is, it is embodied in people’s attitudes towards that brand. It is dynamic, and subject to change over time. It attaches itself to the brand name, providing a current summary of

13

Larry Percy people’s feelings, knowledge and experience with that product or service. Think about chocolate for a minute. Basically, chocolate is chocolate. Or is it? Are some brands better than others? Why? What about washing-up powder? They all get the job done, and use the same basic ingredients. Or do you think some do a better job than others? What about toothpaste, or vodka, or underwear? Where do the differences among brands in these product categories come from? How much of the difference is ‘real’ versus perceived? Why do you prefer one brand over another, especially if when looked at with a coldly objective eye, there is very little, if any, actual difference in the products?

Measuring Brand Equity Brand equity is a result of brand attitude, and this is what provides the key to its understanding. In many ways, building and ensuring a continuing positive brand attitude is what strategic brand management is all about, because it does lead to strong brand equity. The most important thing to understand when you are trying to measure brand equity is that what is needed is a measure of understanding, not a measure of the results or consequences of a brand’s equity. Too often, when people ‘measure’ brand equity, they are really only tracking summary measures of what is going on in the market as a result of the brand’s equity. What is needed is a measure of the components that lead to brand equity, and this means measures of how the market forms current attitudes towards the brand. If we are to really understand a brand’s equity, we must understand how it is constructed. It is this understanding that ensures an effective positioning in our marketing communications, and the ability to adjust that positioning over time as needed to continue building and sustaining positive brand equity. We measure brand attitude using an Expectancy-Value model (considered by most researchers in consumer behaviour to be the best model of attitude). Basically, this model states that a person’s attitude towards something, a brand or product in our case, is the sum of everything they know about it weighted by how important those beliefs are to them. Obviously, we are not able to study ‘everything’ about a

14

Advertising and Brand Equity brand or product, but we can and should consider everything critical to the benefit positioning of the brand. If we are to understand the current equity of a brand, it is necessary to ‘deconstruct’ its positioning in order to access the strengths and weaknesses of the belief structure that sustains people’s attitudes towards it. It should now be clear that to a large extent a brand is not a tangible thing at all, but rather the sum of what someone knows, thinks, and feels about a particular product. In a very real sense, brands only exist in the minds of consumers, but that does not make them any less real. And to a very real extent, brands and the equity attached to them exist as a result of marketing communication, and especially advertising. It is advertising (when successful) that positions a brand in the consumer’s mind, nurtures salience, and builds positive brand attitude that leads to a strong brand equity.

Brand Positioning At its most general, a brand position is a ‘supercommunication’ effect that tells the consumer what the brand is, who it is for, and what it offers. This reflects the relationship between brand positioning and the two core communication effects of brand awareness and brand attitude. It’s easy to understand that one must have strong awareness if a brand is to be considered when the ‘need’ for that type of product (however the consumer defines it) occurs. Strong brand awareness (for almost any brand) must be generated and sustained with marketing communication. It is marketing communication, and advertising in particular, that builds and maintains brand salience. It is not enough for a brand to be recognized if it is to be successful. A brand must occupy a ‘salient’ position within the consumer’s consideration set. In fact, the strength of a brand’s salience is one indicator of the brand’s equity. (A useful measure of this is the ratio of top-of-mind recall to total recall among competitive brands in a category.) Brand attitude, however, is not quite so easy to deal with. Who exactly is the target audience? Is everyone looking for the same thing; or the same things all the time? What is important, and to whom? How are brands seen to deliver on the things important to the target audience? Answers to these questions are critical if we are to positively effect brand attitude.

15

Larry Percy The role of benefits in effective positioning in communication is of course essential. But benefits must be considered in relationship to brand attitude, which in its turn is the link to purchase motive. Consumers hold what we might think of as an overall summary judgement about a brand, following the Expectancy-Value notion of attitude: ‘Hush Puppies makes great shoes’ is an attitude about Hush Puppies that connects the brand in the consumer’s mind with what is the likely purchase motive, sensory gratification (i.e. they buy Hush Puppies to enjoy them). This brand attitude, however, which we might think of as a ‘superbelief,’ doesn’t just spring from nowhere, but is the result of one or more beliefs about the specific benefits the brand is thought by the consumer to offer in support of that overall attitude. Effective communication strategy requires an understanding of what that belief structure is, and how it builds brand attitude. Within the overall positioning that results from this understanding, one we must determine what the benefit emphasis and focus should be (cf. Percy, Rossiter, and Elliott, 2001). To begin with, it is important to remember that purchase motive is really the underlying basis of benefit. Purchase motives are, after all, the fundamental ‘energizers’ of buyer behavior. These same motives also energize the usage of products. Motive-based positioning requires a correct answer to the question of why consumers in the category are really buying particular brands. Unfortunately, most benefits tend to be motivationally ambiguous. One must also be careful to distinguish between motives that drive product category decisions rather than brand decisions. People may buy (say) active casual footwear because they are comfortable (a negative motive), but buy particular brands for more ‘style’ related reasons (a positive motive). This is an absolutely critical distinction. Benefits like comfort or low price relate to negative motives, and are unlikely to drive specific brand purchases. Yet, someone may be looking for a good price in the category, but not at the expense of ‘style.’ The reason this is such an important point is that positive motives suggest marketing communication where the execution itself actually becomes the product benefit. Here more than ever a truly unique execution is required where the brand owns the ‘feeling’ created by the advertisers for the brand. You can’t ‘prove’ you have a

16

Advertising and Brand Equity more ‘stylish’ or popular shoe, but you can make people believe you do.

I-D-U Benefit Emphasis The benefits a brand emphasizes in marketing communication should be selected according to three major considerations: Importance, Delivery, and Uniqueness (cf. Rossiter and Percy, 1997). Importance refers to the relevance of the benefit to the underlying motivation. A benefit assumes importance only if it is instrumental in helping meet the consumer’s purchase motivation. Delivery refers to a brand’s perceived ability to provide the benefit. Uniqueness refers to a brand’s perceived ability to deliver on the benefit relatively better than other brands. What we are looking for are one or two benefits, relevant to the underlying motive, that can produce a perceived difference between alternative brands. These benefits should then be emphasized in the brand’s marketing communication. A note in passing. We are talking about perceived delivery and uniqueness. Just because a brand may not now be perceived to provide benefits that could optimize purchase against important motives does not mean this perception cannot be created (unless, of course, it stretches the consumer’s understanding of the brand, which is one reason we need to fully understand current brand equity).

Benefit Focus The overall positioning of a brand basically chooses a location for the brand in the consumer’s mind. The I-D-U analysis helps one decide which benefit(s) to emphasize. After that, one must decide what aspect of the benefit to concentrate on in the execution of marketing communications. Up to now we have used the term ‘benefit’ in a rather general way. We have considered a benefit as any potential positive or negative reinforcer for a brand, in line with our definition of brand attitude as representing the overall delivery on the underlying motivation. Since a ‘reinforcer’ is anything that tends to increase a response, benefits as we have been talking about them underlie and help increase brand attitude. Now we will distinguish more sharply between attributes, benefits, and emotions, as discussed by Rossiter and Percy (2001).

17

Larry Percy If we think about the underlying motive as ‘why the consumer wants the brand,’ we may consider that • Attributes are ‘what the product has’ • Benefits are ‘what the consumer wants,’ and • Emotions are ‘what the consumer feels.’ A brand, for example may offer attributes that the consumer may or may not think of as a benefit. Benefits, in their turn may have various emotional consequences or antecedents, depending upon the underlying motive. All marketing communication presents or implies a ‘benefit’ as either an attribute, benefit, or emotion as defined above. The key to effective communication is using the appropriate benefit focus. At this point it must seem we are overly complicating things, but this really is a powerful way of ‘fine-tuning’ a positioning, and not nearly as confusing as it may appear. When the benefit focus is not consistent with the underlying purchase motivation, the logic of the message breaks down, and the effectiveness of the communication breaks down. To effectively position a brand, it is necessary to understand what brand attitudes link the brand to the purchase motivation in the consumer’s mind, the proper benefit to emphasize, and how to focus the consumer’s attention on that benefit. Approaching positioning in this way will have a significantly positive effect on building and sustaining brand equity, and in its turn on the success of the brand.

Brand Attitude Strategy Proper positioning encourages a strong link between category need and the brand and it builds positive brand attitude. And as we have seen, it is this positive brand attitude that builds and sustains brand equity. As Rossiter and Percy (1997) define it, effective brand attitude strategy is a function of the level of involvement in the purchase decision (as defined by the level of risk, either psychological or fiscal) and the underlying motivation driving behaviour in the category (negative vs. positive). These two dimensions define the four quadrants of the brand attitude strategy components of their planning grid: low

18

Advertising and Brand Equity involvement/informational, low involvement/transformational, high involvement/informational, and high involvement/transformational (where informational strategies reflect negative motives and transformational strategies reflect positive motives). Picking the correct brand attitude strategy, as defined by the Rossiter-Percy Grid, is essential for effective marketing communication (Rossiter, Percy and Donovan, 1991). This is especially true for advertising built upon transformational strategies. When dealing with positive motives (such as sensory gratification or social approval) the key to an effective execution is emotional authenticity. It must ‘ring true’ to the receiver. While brand equity is more likely to be a function of attributes or benefits when dealing with informational strategies, with transformational strategies brand equity will be strongly influenced by the emotional component of the brand benefit structure. One of the best examples we can offer of this idea of the importance of emotional authenticity in the execution of transformational advertising to effectively build a strong positive brand attitude is advertising for beer in the U.S. market. In the US. market, not only is beer brand equity directly related to the advertising execution, so is the taste of beer! In a fascinating study conducted by the author (along with John Rossiter), beer drinkers were asked to taste and evaluate a number of beers ranging from lower-calorie Miller Lite to premium beers like Budweiser, all the way through malt liquors and Guinness Stout. Represented were beers marketed as ‘regular beers’ (Pabst) vs. premium beers (Budweiser), and ‘lighter’ beers (Miller Lite, Coors) vs. ‘stronger’ beers (Colt 45 Malt Liquor, Guinness Stout). These two dimensions represented the dominant marketing dimensions in the U.S. beer market. In a tightly controlled experiment, subjects tasted the beers presented randomly in clear glass containers, and rinsed their mouths with distilled water and unleavened wafers between tastings. When the results of these taste tests were evaluated and analyzed, the ‘lighter’ beers were separated from the ‘stronger’ beers, and the ‘regular’ beer from the ‘premium’ beer (as we see in Figure 1:1). The tastes of these beers were positioned exactly as their advertising presented them.

19

Larry Percy Figure 1:1

Perceptual Mapping of Beer Taste Evaluations When Brands are Known Budweiser •

Colt 45 • Miller Lite • • Coors

• Guinness Stout

• Pabst

However, a matched set of beer drinkers went through the same exercise, but they tasted the beers blind, not knowing what beers they were tasting. When taste evaluations were analyzed, with the exception of Guinness Stout (which is clearly in Figure 1:2). The brand equity and taste of these beers was wholly defined by their advertising.

20

Advertising and Brand Equity Figure 1:2

Perceptual Mapping of Beer Taste Evaluations When Brands are Blind

Miller Lite

Coors

•• Pabst • • Colt 45 •

Guinness Stout

•

Budweiser

Summary Advertising and brand equity are related in the strongest possible way, as we have shown. Without marketing communication in general, and advertising in particular, there would be little likelihood of any brand awareness, and less of brand salience; the managed development of brand attitude almost impossible. And without both brand awareness and brand attitude there would be no brand equity. It is advertising, effectively positioned to build and nurture a positive brand attitude that leads to the building and maintaining of brand equity.

21

2

Brands are just like real people! The development of SWOCC’s Brand Personality Scale EDITH G. SMIT, EMILIE VAN DEN BERGE & GIEP FRANZEN

Introduction When Kodak is sincere, then IBM can be competent, Marlboro cool and Revlon a little sophisticated. Just like humans, brands can be plain and closed or show their faces and have opinions of their own. Examples are Diesel, Absolut, but also ABN AMRO and Ben. What these brands “say” and “do” have a distinctive style and signature that can be described in terms of personality traits. Brands as human beings: an attractive metaphor. Sometimes they are even positioned as a person, for instance Ben and Ilse. The expectation is that the stronger and more powerful the personality is, the more bonding there can be between brands and consumers. Compare it with the people you know: those with outspoken opinions are interesting. Dull people –in your own perception, of course- are less interesting. Brand personality is not new, and sometimes confused with related brand concepts: brand identity, brand image, brand values, brand loyalty, and user image. Another complicating factor is the measurement of brand personality. In the Netherlands a lot of different approaches and scales are used by practitioners to measure brand personality. A lot of different views and ditto measurement approaches are proposed in the literature. SWOCC started a brand personality project in 2000, in which the concept was explored theoretically and empirically. The theoretical

22

The development of SWOOC’s Brand Personality Scale part resulted in a SWOCC-publication by Marieke van den Berg (2001). The empirical part was published in 2002 by Emilie van den Berge. In this conference paper, we will focus on the development of our measurement scale for brand personality. First, we will briefly introduce the concept. Then we sketch the research program as well as the results. Finally, we will give some examples of results for individual brands.

Brand personality in theory The term brand personality has been borrowed from theories on human personality. Humans are unique and have their own unique personality. Since Aristotle, personality has been studied in many different ways: from a psychoanalytic viewpoint, within behaviorism, from an existential point of view, using the psycho lexical approach and within cognitive psychology (to name only a few different approaches). Exempting differences, the basic view is that personality has something to do with the way people –or better persons- differ from each other. Personality is ‘the thing’ that makes someone a person; it’s the part of an individual that most of the time is consistent and durable. Most of the time, because we interact constantly with persons and objects and differ in different situations (Lippa, 1994). Several attempts have been made to categorize people on the basis of their personalities. Within personality psychology this resulted in the famous “Big Five” (Caprara, Barbaranelli & Guido, 2001; Digman, 1990; Goldberg, 1990; Lippa, 1994; McCrae & John, 1992): Extraversion, Agreeability, Consciousness, Emotional Stability and Openness/Culturedness (see Figure 2:1). A person scoring high on Extraversion could be described as someone who likes to talk (a lot), is active and assertive, energetic and outgoing – instead of quite, shy and silent. Someone scoring low on Agreeableness is described as cold, unfriendly, quarrelsome, while a conscious person is thought to be organized, thorough, efficient and responsible. Not only persons (consumers) are categorized, brands are as well. Brand managers do so in terms of brand and product strategies or brand identities. Consumers do so too, in terms of images they have from brands, experiences and feelings.

23

Edith G. Smit • Emilie van den Berge • Giep Franzen Figure 2:1 The ‘Big Five’ dimensions of human personality Items low on dimension Quiet, Reserved, Shy,

BIG FIVE EXTRAVERSION

Silent, Withdrawn

Items high on dimension Talkative, Active, Assertive, Energetic, Outgoing

Fault-finding, Cold, Unfriendly, Quarrelsome, Hard-hearted

AGREEABLENESS

Sympathetic, Kind, Appreciative, Affectionate, Soft-hearted

Careless, Disorderly, Frivolous, Irresponsible, Slipshod

CONSCIOUSNESS

Organized, Thorough, Planful, Efficient, Responsible

Tense, Anxious, Nervous, Moody, Worrying

EMOTIONAL STABILITY

Stable, Calm, Contented, Unemotional

Commonplace, Narrow interests, Simple, Shallow, Unintelligent

OPENNESS * CULTUREDNESS

Wide interests, Imaginative, Intelligent, Original, Insightful

One of many ways to categorize certain brand attributes is brand personality. Brand personality characteristics are derived directly from people associated with the brand or persons’ own characteristics, and more indirectly via product-related attributes, brand name, logo, communication style, prices and distribution. All resulting in brand personality (Bauer, Mäder & Keller, 2001; Fournier, 1998; Timmerman, 2001). Brand personality, however, has been poorly conceptualized (e.g. Caprara et al., 2001; Patterson, 1999; Van den Berg, 2001). In his literature reviews on brand research until the nineties, Patterson (1999) concluded “Our understanding of brand image has been hampered by poor conceptual development. Although it has been the focus of branding research since 1950s, there continues to be a large degree of confusion about what brand image actually means. In particular, most authors have failed to highlight the distinction between the concepts of brand image, brand personality and user image and, as a consequence, these three concepts have tended to be used interchangeably”. A few examples of brand personality definitions: • “An attitude of mind and tone of voice and set of values” (King, 1973, in Lippa, 1994);

24

The development of SWOOC’s Brand Personality Scale • “The extent to which consumers perceive a brand to possess various human characteristics or traits” (Alt & Griggs, 1988, p.9); • “The way in which a consumer perceives the brand on dimensions that typically capture a person’s personality” (Batra, Lehman & Singh, 1993); • “The personality consists of a unique combination of functional attributes and symbolic values” (Hankinson & Cowking, 1995); • “Brand personality displays the brand’s core characteristics, embodied, described and experienced in human terms” (Restall & Gordon, 1994); • “A brand’s personality … embodies all of the qualities it has to offer over and above its primary characteristics and its functional purpose” (Tennant, 1994, in Patterson, 1999); • “Brand personality reflects customers’ emotional response to a company and its product” (Triplett, 1994); • “The outward ‘face’ of a brand; its tonal characteristics most closely associated with human traits” (Upshaw, 1995); • “The set of human characteristics associated with a brand, which makes it unique, compared to other brands” (Aaker, 1996, p.1); • “The consumer’s emotional response to a brand through which brand attributes are personified and are used to differentiate between alternative offerings” (Patterson, 1999). One of the differences in these definitions is the difference in perspectives: an advertiser versus a consumer view. In an advertiser view, brand personality is an important driver for brand identity (Kapferer, 1996), which is shown by the definitions of Restall and Gordon (‘brand’s core characteristics’) and Tennant (‘the qualities it has to offer’). From a consumer’s perspective, brand personality is seen as one of the components of brand image (Batra, Myers & Aaker, 1996), for instance in the first three definitions: ‘attitude of mind’, ‘consumers perceive a brand to have …’. Although the definitions are different, they are similar in that brand personality is about perception (in the consumer’s view), about personality characteristics attributed to brands, about associations and symbolic values and about emotional responses on the brand or emotional relationships with brands. In our research project, we

25

Edith G. Smit • Emilie van den Berge • Giep Franzen defined brand personality as the set of personality characteristics associated with a brand in comparison with other brands (based on Aaker, 1996).

Brand personality measurement To measure brand personality a lot of different methods are used, in academic as well as commercial research, simply because ‘the’ method does not exist. A widely used (qualitative) method is to ask respondents to imagine a brand as a person (Is it a boy or a girl? Is it young or old?) or an animal (Aaker, 1996; Alt & Griggs, 1988; Fournier, 1998). Also photo-sort techniques are used (Floor & Van Raaij, 1998). In fact, association methods are mostly used, varying from free associations to aided associations (qualitative or quantitative, unstructured or structured). Here, we focus on the Brand Personality Scale of Jennifer Aaker, which is a structured quantitative measurement technique (scaling). The main reason for choosing this kind of technique is that it is standardized and that results for different brands are thus comparable. Aaker’s scale is the most recent fundamentally tested scale found in the literature and not only based on the “Big Five,” but also on other well-defined brand personality scales (e.g. Malhotra’s semantic differential of 1981, Alt and Griggs’ Brand Personality Rating Scale of 1988, Batra et al.’s semantic differential of 1993). Aaker’s procedure – which she described in several articles- can be summarized into six steps: First, Aaker (1996, 1997) collected personality characteristics from psychology (“Big Five” based personality scales; 204 unique traits), from brand practitioners (advertising agencies, scholars and research agencies; 133 unique traits), and from a small-scale qualitative study in which respondents (n=16) were asked to write down characteristics that came to mind when naming six different brands (295 traits). This procedure resulted in 309 non-redundant candidate personality traits. Second, the 309 items were reduced by asking respondents (n=25) to judge the items on a 7-point scale, ranging from (1) not at all descriptive to (7) extremely descriptive. Respondents were asked to think of as many different types of brands as possible. Items with an

26

The development of SWOOC’s Brand Personality Scale average score of 6 and more (very descriptive) were selected, leaving 114 personality traits for the subsequent study. Third, an US representative non-student sample (n=631 respondents from a national mail panel) was asked to rate 37 brands that serve symbolic and utilitarian functions or both on 114 items1 (scale: 1 = not at all descriptive, 5 = extremely descriptive). Factor analysis (Varimax rotation) for the aggregate database (with average scores on the 114 items per brand) showed five dimensions (total explained variance of 92%): Sincerity, Excitement, Competence, Sophistication and Ruggedness. Factor analyses for sub samples (men versus women, younger versus older respondents) also showed these five dimensions, resulting in a stable and robust five-factor-solution. Yet, only three of these brand personality factors were those appearing in the Big Five model for human personality (Caprara et al., 2001). Fourth, for every factor a new factor analysis was done (Aaker, 1997) to identify the traits that most reliably, accurately, and comprehensively represent the five dimensions. This resulted in 15 so called ‘facets’ (see underlined items in Figure 2:2). Each facet was then split into three clusters to select the item with the highest item-to-total correlation (> 0.60). Test-retest correlations (test sample of n=81; r > 0.70) and reliability analyses (Cronbach’s alphas > 0.85) were also used. Based on these steps, the resulting list consisted of 42 items. Fifth, the scale of 42 items was tested again by using a new sample (n=180 respondents of a national mail panel) and a different set of 20 brands. This study confirmed the other results (Confirmatory Factor Analysis with GLS; AGFI = 0.86). Sixth, Aaker repeated her approach in other countries. Based on four studies (Aaker, 1999b; Aaker, Benet-Martinez & Garolera, 2001), a set of brand personality dimensions was revealed that was the same for Japan, Spain and the US: Sincerity, Excitement, Competence and Sophistication (although Competence not as clear in Spain). She also found culture-specific dimensions: an American Ruggedness dimension; a Japanese Peacefulness dimension; and a Spanish Passion dimension. In both non-American countries, the original 42 brand personality items were complemented with culture-specific brand1

Aaker (1996) used primarily positively valenced traits, which were rotated completely.

27

Edith G. Smit • Emilie van den Berge • Giep Franzen Figure 2:2 Brand Personality dimensions (based on Aaker, 1997; 2001)

American Brand Personality dimensions

27%

Sincerity

25%

18%

Excitement

12%

9%

Competence

Sophistication

Ruggedness

Down-to-earth

Daring

Reliable

Upper class

Outdoorsy

Family-oriented

Trendy

Hard working

Glamorous

Masculine

Small-town

Exciting

Secure

Good looking

Western

Honest

Spirited

Intelligent

Charming

Tough

Sincere

Cool

Technical

Feminine

Rugged

Smooth

Real

Young

Corporate

Wholesome

Imaginative

Successful

Original

Unique

Leader

Cheerful

Up-to-date

Confident

Sentimental

Independent

Friendly

Contemporary

related attributes, based on personality scales used in that country as well as association tasks by participants from Japan and Spain. Aaker’s approach was also used by other researchers, for instance by researchers from the Mannheim University in Germany (Bauer et al., 2001), who used the 42 item-scale of Aaker (1997) for measuring brand personality for seven brands in Germany and Spain. This study, however, did not find the five factor solution of Aaker, maybe because they used convenient samples (i.e. students) and did not –as Aaker didcomplement their set of personality items with culture-specific items.

Overview research project The aim of our research project was not to replicate Aaker’s studies. Our aim was to have a clear understanding of brand personality, to sketch the usefulness of the concept for brand strategy, and especially to develop a standard method (i.e. scale) for measuring brand personality. Within the scope of SWOCC, we aimed to develop a scale –in a fundamental way– that is useful for practitioners.

28

The development of SWOOC’s Brand Personality Scale The empirical part of our project consisted of two phases. In the first phase, we developed and measured our scale for 20 brands in four product categories. Based on these results, we found seven dimensions of brand personality. In the second phase, we used the same procedure and questionnaire for 73 extra brands in eight product categories. The aim of this second phase was to validate our seven dimensions and to have a more solid base (in terms of number of brands) for the reduction of items.

Selection of Personality items First, we translated Aaker’s Brand Personality Scale of 42 items into Dutch. However, the problem with our language is that most American terms have more than one equivalent in Dutch. Therefore, seven persons, who spoke English as well as Dutch fluently were asked to translate the Aaker list into Dutch. We selected the items which were most agreed upon. Second, we added a list of Dutch personality items because research by Aaker (1999b), Aaker et al. (2001) and Bauer et al. (2001) showed that not all of Aaker’s American dimensions were found in other cultures (Japan, Spain and Germany). We used items from the Dutch Personality Scale by Brokken (1978), who based his list on the standard Dutch dictionary and a study in which 200 couples were asked to judge one another or themselves. These judgments eventually resulted in 1203 items and six factors. Like the “Big Five”, Brokken found dimensions expressing a sympathetic factor (agreeableness), emotional factors (stability and extraversion) and ‘organized’ factors (consciousness or orderliness/preciseness). As a third step, we reduced the Brokken list by removing ‘oldfashioned’ words (his dissertation was published in 1978) and removed items with factor loadings lesser than |0,35|. We then asked four experts in the field of brand research and six non-experts (consumers) to go through the remaining list and select those items that can be descriptive for a brand. They were asked to think of as many different brands as possible, not one specific brand. Finally, we based the selection on the items that were mostly agreed upon (no differences were found between experts and others). This resulted in 60 Brokken

29

Edith G. Smit • Emilie van den Berge • Giep Franzen and 42 Aaker items. Appendix A shows our list of 102 personality items.

Selection of Brands We started with 20 different brands in four product categories (see Table 2:1). We selected salient well-known brands from different categories: cars, beer, shampoo and tele services. After this phase, several organizations were willing to finance the measurement of their brand(s) according to our procedure (with the 102 items). This gave us the opportunity to enlarge our sample of brands to a total of 93 brands in 11 product categories. The only ‘disadvantage’ of these extra measurements was that some brands were –on request- measured more than once (see * in Table 2:1). Table 2:1 Measured brands Product category Brands Cars Beer Shampoo Tele Services Beverages Cigarettes

Rolling tobacco Financial banks Mail carriers Insurances Tele Services Shops

Phase I BMW, Volvo, Mazda Heineken, Grolsch, Amstel, Bavaria, Palm, Dommelsch Andrélon, Nivea, Head & Shoulders, Pantène Pro-V, Sanex, Organics KPN, Libertel, Telfort, Dutchtone, Ben Phase II Douwe Egberts *, Nescafé, Max Havelaar, Coca Cola *, Spa *, Cup a Soup, Pickwick, Lipton Ice Tea, Appelsientje Gauloises Blondes, Caballero, Barclay, Lucky Strike, Marlboro *, Philip Morris *, Chesterfield *, Camel *, Mild Seven * Drum, Van Nelle, Samson, Brandaris, Javaanse Jongens, Jacobs ING Bank, Postbank, ABN AMRO, Rabobank PTT Post, Deutsche Post, Royal Mail, UPS, TNT Delta Lloyd *, Nationale Nederlanden *, Aegon *, Centraal Beheer *, Amev *, Ohra, Interpolis Libertel, Dutchtone, KPN, Ben (* all in phase I as well) Albert Heijn, Konmar, C1000, Aldi, Bijenkorf, Hema

Note: * = measured more than once

Method Because we had two phases of data gathering, our measurements were done in two online panels. By email the respondents were asked to

30

The development of SWOOC’s Brand Personality Scale complete the questionnaire for one product category with a maximum of six brands per questionnaire (i.e. respondent). In the first phase (summer 2001), we used the CentERpanel of the University of Tilburg (the Netherlands), which is a so-called ‘real’ household panel with a firm number of panel members. To avoid the possibility that someone would answer the questionnaire again, we used another online panel in the second phase (February 2002), namely the online Internet panel of Interview/NSS. This second panel differed from the first one, because it was a so-called ‘occasional’ panel with thousands of respondents known to possess certain characteristics and willing to complete questionnaires on a regular basis. The advantage was the possibility to ‘select’ a lot of sub samples for the different product categories while, at the same time, exerting some control over the composition of the sub samples (in terms of sex, age and education).

Questionnaire Brand personality was measured with ten matrices. Each matrix showed the brands and the option “none of these brands” in the columns (between three and six brands per matrix) and the characteristics in the rows (about 10 items per matrix). The rating was done by means of ASSPAT, which stands for ASSociation PATtern method. In ASSPAT, respondents were asked to compare objects (brands in this study) for different items. The respondent was asked to mark one or more brands when he or she finds an item descriptive for the brand(s). It was also possible to mark “none of these brands”. In other words, brands were scored (per item) on a dichotomous scale (0 = not descriptive, 1 = descriptive). Although the richness of data is less than would be the case for an interval or ordinal scale (like Aaker used), the great advantage of this method is that 102 items and several brands can be measured in one questionnaire without exhausting the respondent. This is especially important in online measurement. The items were offered randomly, in other words: there was no fixed sequence in matrices or items within a matrix. Besides brand personality, we measured experience with products or services, most used brands in the category and likeability of the brands (school mark, rating from 1 tot 10). The demographics were already

31

Edith G. Smit • Emilie van den Berge • Giep Franzen known since we worked with online panels. In the first phase, we combined the scores with scores of the respondents on themselves (which were already known per panel member). In the second phase, we added questions about the ideal brand in the measured product category (see Discussion).

Response and Sample In the first phase, we worked with a total net sample of 1009 respondents (1294 members were asked to complete a questionnaire of which 78% responded). In the second phase, 3524 of the 9650 panel members completed the questionnaire (response of 37%). When comparing sample characteristics (sex, age and education), the second sample was more representative for the Dutch population than the first one. This was due to the selection process in the second panel (quota sampling). The first panel consisted of more men, higher educated respondents and more respondents in the age of 31-55. Although not representative for the Dutch population, the sample reflected the Internet population (mean age sample I = 46.5, mean age sample II = 40). Table 2:2 Characteristics, two samples (column %) Sample (n=1009) Men 55.4 Women 44.6 Low education 14.3 Medium education 30.0 High education 55.7 Age: 13-19 0.8 Age: 20-30 12.0 Age: 31-40 25.9 Age: 41-55 35.7 Age: 56+ 25.5 Source population: Intomart/GfK 2002, January

32

I Sample (n=3524) 49.8 50.2 9.7 46.2 44.1 9.4 17.8 20.2 26.4 26.2

II Population 48.9 51.1 38.1 33.9 28.0 9.4 17.7 20.0 26.5 26.4

The development of SWOOC’s Brand Personality Scale

Results Dimensions First, we factor analyzed the 42 Aaker-items (see * in appendix A), at the respondent as well as the aggregate or brand level. As Aaker, we used Principal Components Analysis with Varimax rotation based on the correlation matrix. Based on data of the first phase (20 brands), the exploratory factor analysis (EV>1) resulted in six factors (with cumulative R2 of 92%). Only two of these six factors corresponded with Aaker’s dimensions, namely ‘Sincerity’ and ‘Excitement’. Although the factor analysis did not show the Aaker-dimensions in our study, the reliability of the proposed items per Aaker-dimension was high (Cronbach’s alpha per Aaker dimension was between 0.81 and 0.98). We repeated the analyses for all 102 items (the Aaker items plus the extra Dutch items). Factor analyses resulted in seven factors (explained variance of 93%, EV>1), again with the two factors ‘Sincerity’ and ‘Excitement’. Second, we repeated our analyses on the complete dataset of two waves (93 brands), after we checked for panel differences2. The exploratory factor analysis (Principal Components with Varimax rotation, based on correlation matrix, EV > 1) resulted in 10 factors of which the last three did not have distinctive factor loadings (also shown in the screen plot). The factor analysis for seven factors resulted in the dimensions listed in Table 2:3, which were more clear and distinctive than it was the case for 20 brands (although the cumulative R2 is a little less, namely 87%). In Table 2:3, the results of the factor analysis are listed. Items that scored relatively high on the factor (factor loadings ≥0.60)3 and relatively low on the other dimensions 2

3

T-tests with average scores on the items as dependent variables and panel, as group variables showed no significant differences (p > 0.05), except for the three items ‘decent’ (t(df=91)=2.15, p=0.03), ‘exciting’ (t(df=91)=2.33, p=0.02), and ‘unfriendly’ (t(df=91)=-2.39, p=0.02). The items ‘vital’, ‘western’, ‘rebellious’, ‘good looking’ (factor 2), ‘vulnerable’ (factor 3), ‘radical’, ‘outdoorsy’ (factor 4) and ‘uncomplicated’ (factor 7) are excluded because of their factor loadings (0.60) and low on the other dimensions are shown; bold = factor name; underlined = facet name (based on highest item-to-total correlation per facet); * = Aaker-items

35

Edith G. Smit • Emilie van den Berge • Giep Franzen (items: firm and stabile). This dimension is actually a combination of Aaker’s Competence and Sincerity dimensions – that is why we labeled it ‘Competence plus’. Our dimension consisted of 16 Aaker items, of which 12 Competence-items (combining our facets confident, accurate and respectable) and 4 Sincerity-items (especially our facet sympathetic). With 33% explained variance, Competence-plus is our strongest dimension. Examples of brands that score relatively high on Competence+ are: some car brands (Volvo, BMW), and some financial institutions (ABN AMRO, Postbank and Rabobank). Examples of brands scoring very low on the Competence+ dimension (in other words are not found to be described by items belonging to this factor) are different cigarette and rolling tobacco brands, and some shampoo brands (Pantene Pro-V and Organics).

2. Excitement+ The Excitement-dimension with 22% explained variance is more or less the same as Aaker’s dimension (5 of her Excitement items were grouped here), although broader. Our dimension combines items such as Cheerful (Jolly, Happy, Spontaneous and Enthusiastic), Spirited (Active) and Imaginative (Creative and Original). Also in this dimension, we found a small part of Aaker’s Sincerity dimension (her Cheerful facet). Relatively high scoring brands are: beverages (Coca Cola, Lipton Ice Tea, Grolsch), but also for instance the two tele service brands Libertel and Ben. ‘Low’ scoring brands are cigarette and rolling tobacco brands as well as the insurance brands Amev, Ohra and Delta Lloyd.

3. Gentle With 10% explained variance, Gentle is our third one-facet-dimension with no resemblance to one of Aaker’s dimensions. It does, however, show some resemblance with the “Big Five” dimension Agreeableness. The dimension is best described by the items softhearted, kind, feminine and amiable.

36

The development of SWOOC’s Brand Personality Scale Beverages is a product category that scores relatively high on Gentle. Examples of Gentle-brands are: Pickwick, Appelsientje and Spa. Also Nivea and Mazda are seen as ‘gentle’. On the low side, we found some not Gentle brands such as Coca Cola and Head & Shoulders. Our last four factors are not very strong (30%) Javaanse Jongens, Jacobs, Brandaris, Samson, Royal Mail, Telfort, Dutchtone, Ben, Pantene Pro-V, Organics (all < 10%) Ben, Libertel, Coca Cola, Jacobs, Brandaris, Samson, Lipton Ice Tea, Cup a Soup, Van Nelle, Mild Seven, Grolsch, Albert Heijn, Chesterfield, Barclay, Philip Bijenkorf, Hema, BMW, Morris, Gauloises Blondes, Mazda, Centraal Beheer, Caballero, Lucky Strike, Postbank, PTT Post (all > Royal Mail, Deutsche Post, 20%) Head & Shoulders, Nivea, Amev, Ohra, Delta Lloyd, Max Havelaar (all < 10%)

The development of SWOOC’s Brand Personality Scale Pickwick, Appelsientje, Spa, Deutsche Post, UPS, Royal Nivea, Mazda, Mild Seven Mail, TNT, Head & (all > 20%) Shoulders, Caballero, Lucky Strike, Marlboro, Philip Morris, Brandaris Coca Cola, Telfort, Delta Lloyd, Aegon (all < 5%) BMW, Volvo, Coca Cola, Mild Seven, Barclay, Ruggedness Chesterfield, Jacobs, M=11, SD=7 Grolsch, Marlboro (all > 20%) Appelsientje, Pickwick, Spa, Royal Mail, Nivea, Pantene Pro-V, Andrelon, Organics, Mazda (all < 5%) Aldi, Konmar, Ben, KPN, Sanex, Organics, Pantene, Annoying Dutchtone, Rabobank, Nivea Pro-V, Andrelon, M=8, SD=3 Postbank, ABN AMRO, Royal Mail, UPS, TNT, BMW, Mild Seven, Jacobs, Deutsche Post, Douwe Head & Shoulders, PTT Egberts, Spa, Palm, Grolsch, Post (all > 11%) Volvo, Bijenkorf (all < 5%) Distinguishing Ben, Max Havelaar, Palm, Nivea, Andrelon, Sanex, Grolsch, Bijenkorf, Lipton Head & Shoulders, M=8, SD=5 Ice Tea, BMW, Aldi, Deutsche Post, Royal Mail, Centraal Beheer (all > 15%) Philip Morris, Mild Seven, Chesterfield, Aegon, Samson, Jacobs, Telfort (all < 5%)

Gentle M=10, SD=6

Note: M=mean (%), SD=standard deviation, average brand score means that x % of the respondents thought that dimension to be descriptive for that specific brand

43

3

Behavioural Finance-based advertising research in the mutual fund industry KLAUS PETER KAAS & JENNY JORDAN

Advertising in the Mutual Fund Industry In the last years, advertising expenditure within the mutual fund industry has increased significantly. In Germany, it rose to 145.61 million Euro in 2001, which is more than twice as high as three years earlier (66.75 million Euro, AC Nielsen 2002). This development was caused by a promising market potential and fierce competition in the investment industry, which in turn was the result of the internationalisation of financial markets, technological changes, and fundamental changes in the investment behaviour of private households. Obviously, advertising has become an important marketing instrument in the financial services industry (Meidan, 1996; McKechnie & Leather, 1998). But what do we know about its effectiveness? Until now, advertising research has neglected industryspecific product characteristics in the investment market and their relevance for advertising effectiveness. There is no doubt that many theoretical and empirical findings of behavioural advertising research also apply to investment products, among these the creation of awareness or emotional experiences through advertising. There are, however, special features of investment products which advertising research should take explicitly into account. Like all investment products, mutual funds are intangible, complex and very homogenous products, the customer benefit of which originates in abstract monetary values (Meidan, 1996; McKechnie, 1997; Sirri/Tufano, 1993).

44

Advertising research in the mutual fund industry Moreover, investment decisions are characterised by high exogenous uncertainty, as future product performance must be estimated on the basis of a set of noisy and vague variables. Consequently, investors’ expectations to uncertain future events play a crucial role in investment decision-making (Wärneryd, 1999). Most importantly, purchase decisions in investment markets follow two dominant criteria: perceived investment risk and expected return (Ganzach, 2000; Olsen, 1997), constructs which apply exclusively to the financial services industry. Risk and return are crucial variables in financial decisionmaking, as indicated in the fundamental normative model of investment behaviour, the mean-variance portfolio analysis (Markowitz, 1952; Tobin, 1958; Sharpe 1994). Marketing research should detect how investors’ perception of these product-specific decision criteria is influenced by marketing instruments, e.g. advertising stimuli. This paper describes a new theoretical approach to advertising in the mutual fund industry and presents empirical insights regarding the influence of advertising on private investors’ risk-return perceptions. Hypotheses are tested by means of an extensive experimental study, and practical implications are described in the last section of the paper.

Behavioural Finance-Based Advertising Research: RiskReturn-Perceptions as Variables of Advertising Effects Behavioural Finance, a field of research that interfaces with economics, finance and psychology, is a relatively young paradigm, which was developed in the US in the late 80‘s because of increasing empirical evidence that existing financial theories were deficient in a real market setting (Shefrin, 2000; De Bondt, 1998; Kahneman & Riepe, 1998). Contrary to the normative approach of classical portfolio theory, Behavioural Finance deals with the descriptive analysis of actual behaviour of individuals in financial markets and analyses psychological influences on information processing and financial decision-making. Risk perception and return estimations are considered to be crucial constructs in the context of financial decisionmaking (Ganzach, 2000; MacGregor et al., 1999; Olsen, 1997; Antonides & van der Sar, 1990). However, traditional behavioural advertising research focuses on rather general categories of advertising 45

Klaus Peter Kaas • Jenny Jordan effects, like awareness, recall or attitude change (Meyers-Levy & Malaviya, 1999; Vakratsas & Ambler, 1999). In this paper, private investors’ risk-return perceptions are considered special variables of advertising effectiveness in the mutual fund industry. How does advertising influence private investors’ perceived risk and expected return? Answering this question leads to a product-specific and selective analysis of the persuasive impact of advertising in the investment industry. The notion of risk refers to the uncertainty of future returns. In classical finance theory, risk is interpreted in terms of variations of returns and measured as variance or betas (Olsen & Cox, 2001; Spremann, 2000). Investment risk is considered a precise, abstract and purely technical statistical concept. This risk concept, however, does not reflect private investors’ understanding of risk. In particular, private investors have a more intuitive, less quantitative, rather emotionally driven risk perception (Olsen & Cox, 2001). Empirical studies that deal with investors’ risk perceptions (Olsen & Cox, 2001; MacGregor et al., 1999; Olsen, 1997; Holtgrave & Weber, 1993) detect four different dimensions of perceived risk: •

• • •

Downside Risk: The risk of suffering financial losses due to negative deviations of returns, starting from an individual reference point. Upside Risk: The chance of realising higher-than-average returns, starting from an individual reference point. Volatility: The fluctuations of returns over time. Ambiguity: The subjective feeling of uncertainty due to lack of information and lack of competence.

All these different aspects have to be taken into account, as single item-measures lead to an incomplete and simplified measurement of the perceived risk construct (Farrelly & Reichenstein, 1984). Expected return, on the other hand, is a simpler, one-dimensional numerical construct, which can be measured in absolute or relative terms or in intuitively quantitative evaluations (MacGregor et al., 1999). Private investors’ risk-return perceptions are always subjective in nature and marked by situational influences. Cognitive aspects as well as emotional states determine their formation (Olsen & Cox, 2001; Ganzach, 2000; Geist, 1999). Therefore, informative and emotional

46

Advertising research in the mutual fund industry stimuli in the ad will influence risk-return perceptions. Depending on the way information and emotions are communicated in the ad, investors will form different risk-return perceptions with regard to the advertised mutual fund.

Relevance of Private Investors’ Judgmental Heuristics for Risk-Return Perceptions and Advertising Content Judgmental heuristics used by investors during information perception and information processing are a field of special interest in Behavioural Finance theory (Kahneman & Riepe, 1998; Tversky & Kahneman, 1974). Behavioural Finance looks at the investor as a ‘homo heuristicus’, using judgmental heuristics instead of formal statistical analysis in information processing and decision-making. Judgmental heuristics are abridged, often sub-optimal information processing strategies, so-called ‘mental shortcuts’ or ‘rules of thumb’, which are used systematically but often unconsciously to simplify decision-making (Kahneman & Tversky, 1996). Heuristics are of special relevance in situations with uncertain or incomplete information and whenever quantitative forecasts, predictions or evaluations of uncertain future events have to be made (Tversky & Kahneman, 1974). As a result of this decision context, cognitive and emotional heuristics produce systematic bias in risk-return perceptions (Raghubir & Das, 1999; Kahneman & Riepe, 1998; Evensky, 1997). These heuristics have to be taken into account when assessing advertising effectiveness. In the next sections, we will give a brief description of two fundamental cognitive heuristics and one affect-based heuristic and develop hypotheses for advertising content and advertising effectiveness.

Anchoring Heuristic When making forecasts, predictions or probability assessments, people tend to rely on a numerical anchor value that is explicitly or implicitly presented to them (Chapman & Johnson, 1999, 1994; Kahneman & Tversky, 1974). Anchoring effects are not restricted to numerical values that are logically coherent with the subsequent numeric

47

Klaus Peter Kaas • Jenny Jordan estimate. According to the so-called “Basic Anchoring Effect” (Wilson et al., 1996), any random and uninformative starting point might represent an initial anchor value which leads to bias in forecasts and estimates towards the value of that initial starting point. Anchoring effects have been identified in many empirical studies and in various decision fields (Mussweiler & Strack, 2001). This robust judgmental heuristic is of particular relevance in financial markets, where it applies to financial forecasts (e.g., stock market prices), leading to severe bias (Stephan, 1999). In practice, numerical data play an important role for the informative content of mutual fund ads. Almost every print ad and many TV-spots highlight figures like past performance data, assets under management, distribution of dividends etc. In addition to direct anchor values (e.g., “10 %”, “10 billion Euro”), indirect and uninformative anchor values with dimensions other than return or monetary units (e.g., “15,000 research specialists worldwide”, “1,000 dreams come true”) can also exert an influence on estimates. In accordance with the anchoring heuristic, these figures will distort return perceptions towards the anchor value. H 1: A low anchor value in an ad will lead to a lower return estimation, compared to a high anchor value.

Representativeness Heuristic People tend to rely on stereotypes. They judge the likelihood of an event according to how it fits into a previously established schema or mental model. They consistently judge the event appearing to be the more representative to be the more likely, without considering the prior probability, or base-rate frequency of the outcomes (Kahneman & Tversky, 1973; 1972). Representativeness is a commonly used and very problematic heuristic in financial markets, as it leads to misinterpretation of empirical or causal coherence (Fisher & Statman, 2000; De Bondt & Thaler, 1993). Illusory correlation and bias in the use of judgment criteria are typical consequences. For instance, past performance data and trend patterns of mutual funds’ performance charts are extrapolated into the future without considering the exogenous uncertainty and randomness of financial markets (Hilton,

48

Advertising research in the mutual fund industry 2001; Raghubir & Das, 1999; Moore & Kurtzberg, 1999; Hulbert, 1999). In terms of practice, mutual fund ads suggestively promote stereotype thinking by communicating positive past performance data, rankings and awards, and by pointing out specific brand values like trustworthiness, competence and experience. Due to stereotype thinking (brand associations, brand schemata), private investors’ riskreturn perceptions will depend heavily on the investment company behind the investment product. H 2: A well-known investment company with a clearly positive brand image will evoke better risk-return perceptions at the product level than an unknown investment company without a clear and positive brand image profile, even though both companies advertise identical products and provide identical product information.

Affect Heuristic Modern financial theory increasingly recognises the fact that financial decision-making is also determined by affective states (Shefrin, 2000; MacGregor et al., 2000). Negative emotions such as fear, worry, anger or shame, and positively experienced emotions like hope, greed, pleasure and joy may influence risk-return perceptions and investment behaviour (Shefrin, 2000; Geist, 1999; Young & O’Neill, 1992). A direct influence of emotions on risk perception and expected returns can be deduced from the affect heuristic (Slovic, 2001; Finucane, Alhakami, Slovic & Johnson, 2000), which assumes that perceptions of risks and benefits related to an alternative are derived from global affective evaluations and associations. If a stimulus arouses a positive affective impression, the decision-maker will judge the risks related to this alternative to be lower and the benefits to be higher, compared to neutral emotional states. If a stimulus is associated with negative affective impressions, the opposite effect will occur: risks are judged to be higher; whereas the benefits are judged to be lower. In practice, mutual fund ads most often contain emotional pictures and emotional slogans in addition to product information. In terms of the affect heuristic, these emotional elements exert a direct influence on investors’ risk-return perceptions.

49

Klaus Peter Kaas • Jenny Jordan H 3: If the emotional content in the ad (pictures, slogans, tonality) succeeds in evoking positive affective impressions towards the advertised mutual fund, the investor will judge the investment risk to be lower and the return to be higher, compared to a purely informative ad.

The Moderating Impact of Private Investors’ Expertise It is important to discuss possible moderating factors in the use of heuristics. Do only inexperienced, unmotivated and relatively incompetent investors use these heuristics, or are they applied by novice and expert investors as well? The question whether or not knowledge influences heuristic information processing has been a topic of discussion. Some researchers propose the unconsciousness and automatism of judgmental heuristics, implying that both lay and expert investors make systematic use of them (Tversky & Kahneman, 1974). Indeed, some empirical findings reveal that investors’ expertise has no influence on the use of judgmental heuristics (Northcraft & Neale, 1987; Joyce & Biddle, 1981; Stephan, 1999). Others, however, demonstrate the moderating role of individuals’ knowledge, stating that knowledgeable persons do not, or only to a moderate extent, apply judgmental heuristics (Mussweiler & Strack, 2000; Wilson et al., 1996). H 4: Less knowledgeable investors make use of the judgmental heuristics to a larger extent than more knowledgeable and experienced investors.

Experimental Study Method Design The hypotheses were explored with a large-scale laboratory experiment between subjects and a 2 (anchor height: low vs. high) x 2 (emotional content: emotional vs. non-emotional ad) x 2 (brand logo: “Fidelity” vs. “DWS”) x 2 (investors’ expertise: competent vs. incompetent investors) factorial design (see Table 3:1). The different graphical elements of the ad represented the experimental factors and

50

Advertising research in the mutual fund industry were systematically varied to test for possible interaction effects. Investors’ expertise as a separate experimental factor was controlled by ex post-blocking of the experimental groups. In accordance with their subjective self-assessment of their knowledge and experience in mutual funds and investment products, participants were divided into groups of incompetent and competent investors after the experiment. Table 3:1 Experimental Design emotional ad

brand: „Fidelity“ high anchor lLow anchor competent incompetent competent investors investors investors

brand: „DWS“ high anchor low anchor incompetent competent incompetent competent incompetent investors investors investors investors investors

n=35 n=28 high anchor competent incompetent investors investors

n=37 low anchor competent investors

n=36

n=35 high anchor incompetent competent investors investors

n=30

nonemotional ad

n=28

n=34

n=36

n=27

n=27

n=31

n=38 n=21 low anchor incompetent competent incompetent investors investors investors n=25

n=31

n=499

Stimuli The advertising stimulus was a print ad promoting a European large and mid-cap mutual fund. Each experimental group was presented with a different version of the print ad representing the systematically varied experimental factors, in accordance with the experimental design. Apart from those experimentally manipulated factors, all ads were identical, containing the same product information and communicating the same risk-return profile of the mutual fund. The anchor value in the ad was an uninformative, irrelevant numerical value, according to the “basic anchoring effect” (Wilson et al., 1996), and was included in the ad as an integral part of the mutual fund’s name (low anchor: “Euro Star 100 FundTM”, high anchor: “Euro Star 500 FundTM”). The mutual fund’s name was a prominent element of the ad. Besides product information, the emotional ad included an emotional image and an emotional claim. The non-emotional ad, on the other hand, did not contain any emotional elements, only emotionally neutral visual background elements and claims. To test for representativeness effects, one half of the ads contained the brand logo of a large, well-known German investment company (“DWS”) and the

51

Klaus Peter Kaas • Jenny Jordan other half the brand logo of a large, but in Germany widely unknown American investment company (“Fidelity Investments”). Participants and Procedure The experimental study was carried out with business students at Frankfurt University, Germany, during December 2001. 526 students of various levels and courses participated in the study (27 incomplete questionnaires had to be excluded from the analysis). After watching the print ad, they were asked to fill in a questionnaire containing the following dependent variables:

• •

Measures Perceived risk: four items measuring the different aspects of the construct (downside risk, upside risk, volatility, ambiguity) Expected return: participants had to estimate the average rate of return per year in the following years

Additionally, participants’ subjective knowledge and experience with regard to investment decisions (four items) were measured via selfassessment. Questions capturing brand image, brand awareness and emotional evaluation of the mutual fund completed the questionnaire in order to comprehend and reconstruct relevant heuristic influences. The most important scales are shown in the appendix.

Results The impact of the anchoring heuristic In case of the low anchor value („Euro Star 100 FundTM“) mean expected return estimation was 11.8 %, in case of the high anchor value („Euro Star 500 FundTM“) 22.6 %. As these results indicate, return evaluations are significantly distorted by anchor size. An analysis of variance revealed a significant main effect (F(1, 483)= 122,371, p < 0.01) with strong effect size1 (eta2 = 0.202). These results demonstrate the “basic anchoring effect” (Wilson et al., 1996): Uninformative and implicitly presented anchor values without logical 1 In accordance with Cohen’s (1988) classification of effect sizes, we assessed effect sizes of eta2 = 0.01 as small, 0.06 as mean and 0.14 as strong effect sizes.

52

Advertising research in the mutual fund industry coherence to the estimate are able to bias return perceptions significantly. The impact of the representativeness heuristic The use of different brand logos („Fidelity“ vs. „DWS“) in the print ad and, correspondingly, the evocation of varying brand associations, have no significant influence on expected return. The direct numeric estimate is obviously not affected by variations in associated brand image. However, there was a significant (F(1, 483)=530,956, p < 0.01) and large (eta2= 0.524) main effect of brand logo on perceived investment risk: Integration of the brand logo “Fidelity” leads to a mean factor score of perceived risk of 0.63, the logo “DWS” leads to a mean factor score of -0.634. This means that the participants perceived investment risk as lower when the mutual fund was offered by the brand “DWS”, although the same product information and risk-returnprofile was communicated. Apparently, thinking in stereotypes and brand schemata leads to bias in risk perception. Further analyses clarify the functionality of representativeness: The image2 of the brand „DWS“ was judged better (mean: 26.8) than the brand image of „Fidelity“(mean: 18.7), t=-16.291, p < 0.01. Similarly, the brand awareness of “DWS” is higher (mean: 4.3) than that of „Fidelity“ (mean: 2.4), t=-18.038, p< 0.01. To sum up, thinking in stereotypes, combined with a higher awareness of the corporate brand, affects perceived investment risk at the product level: A good brand image and high brand awareness result in a lower perception of investment risk. The impact of affect heuristic The emotional content of the print ad had no significant influence on expected return. Apparently, direct numeric estimates are not distorted by emotional elements in the ad. But again, analysis of variance revealed a significant (F(1, 483)=274,107, p < 0.01) and strong (eta2= 0.362) main effect of emotional content on risk perception: The emotional ads resulted in a mean factor score of perceived risk of – 0.456, the non-emotional ads, on the other hand, led to a mean factor 2 Brand image was measured by means of seven items. As Cronbach’s alpha was high (.92), items were summed up to form a brand image score.

53

Klaus Peter Kaas • Jenny Jordan score of 0.452. Thus emotional elements in the ad evoke a decrease in investment risk perception. Further analyses illustrate these emotional effects: Participants’ affective evaluation of the advertised mutual fund (“I have a good feeling with regard to this mutual fund”) is better if emotional elements are present in the ad (mean: 3.3, five point scale) compared to a purely informative ad (mean: 2.1), t=16.046, p < 0.01. In conclusion, positive emotional associations caused by emotional advertising decrease perceived investment risk The moderating impact of participants’ expertise An analysis of variance revealed a significant (F(1, 483)=10,288, p < 0.01) but small (eta2= 0.021) main effect of participants’ expertise3 on expected return: On average, participants with low knowledge and experience had an estimated return of 18.8 %, while competent participants had a higher estimated return of 15.6 %. Furthermore, there was a significant (F (1, 483)=22,567, p < 0.01) but small (eta2= 0.045) main effect of participants’ competence on perceived risk. Competent participants show lower risk perceptions than incompetent participants (mean factor score of incompetent participants: 0.128, of knowledgeable participants: -0.132). This result is probably due to the perceived ambiguity in the evaluation of risky alternatives. This special aspect of risk perception (perceived risk item no. 2) is directly related to participants’ subjective knowledge and available information. The degree of perceived ambiguity differs clearly between incompetent and competent participants, as separate analyses of variance on each aspect of perceived risk reveal: Competent participants experience lower ambiguity (mean: 2.5) in the decision context than their incompetent colleagues (mean: 3.3); F (1, 483)=88,327, p < 0.01, eta2= 0.155 (large effect). Table 3:2 shows the interaction effects between participants’ expertise and the use of heuristics.

3 The four items which measured knowledge and experience related to one factor labelled „competence“. Factor scores were calculated and competence ranks were constructed (2 percentiles): One group was formed for incompetent participants, one for competent participants. Participants’ competence was treated as a separate experimental factor in the factorial design.

54

Advertising research in the mutual fund industry Table 3:2 Interaction Effects between Use of Heuristics and Investors’ Competence Interaction Effect Emotion/Competence

Interaction Effect Anchor/Competence

,8

24

,6

expected return

22 20 18

anchor height

16 14

low

12 10

incompetent

high competent

Perceived Investment Risk

26

,4

,2

-,0

emotional content

-,2

emotional -,4

non-emotional competent

-,6

incompetent

Interaction Effect Brand/Competence Perceived Investment Risk

1,5

Interaction Effects

1,0

,5

Logo (brand)

0,0

Fidelity

-,5

-1,0

incompetent

Interaction Effect

Dependent Variable

Significance

Effect Size

Anchor/Competence

Expected Return

p < 0,05

e2 = 0,008 (small)

Emotion/Competence

Perceived Risk

p < 0,01

e2 = 0,023 (small)

Brand/Competence

Perceived Risk

p < 0,01

e2 = 0,097 (medium)

DWS competent

As an analysis of variance reveals, the biasing effect of the anchoring heuristic is larger for incompetent participants than for more knowledgeable participants. This result is very informative, as it reveals that both competent and incompetent investors use heuristics, but incompetent investors show a bias which is larger in magnitude. This magnitude effect also applies to the affect heuristic. The biasing effect of emotional content on risk perception is larger for incompetent participants than for competent participants. Furthermore, thinking in stereotypes and brand schemata leads to more severe effects on risk perception in the case of incompetent investors. The difference in risk perception due to integration of various brand logos is larger for incompetent participants. Further analyses reveal that competent participants perceive differences in brand image between „Fidelity“ (mean: 21.5) and „DWS“ (mean: 27.0), t=-7.736, p < 0.01, as lower than incompetent participants (mean “Fidelity”: 16.2, mean “DWS”: 55

Klaus Peter Kaas • Jenny Jordan 26.5), t=-16.857, p < 0.01). Similar effects can be found for differences in brand awareness: Brand awareness differences between “Fidelity” and “DWS” are smaller for competent participants (mean “Fidelity”: 3.1, mean “DWS”: 4.6, t=-9.854, p < 0.01) than for incompetent participants (mean “Fidelity”: 1.8, mean “DWS”: 4.0, t=-18.565, p < 0.01) As a consequence, these differences in brand associations and brand awareness lead to varying effects on risk perception. These results support our findings that although both competent and incompetent participants make use of heuristics, the bias is larger for incompetent participants. Apart from these interaction effects between participants’ expertise and the use of heuristics, no other interaction effects were found.

Conclusion •

• • •

Anchoring, representativeness and affect heuristic are used by private investors in an advertising context and lead to bias in risk return judgments, with the following restrictions: Anchoring effects only occur in case of expected return, i.e. direct numeric estimates. Affect and representativeness heuristic only affect perceived investment risk. The referred judgmental heuristics are found in both groups of competent and incompetent investors, but the latter show larger bias than competent investors.

Discussion Advertising in the mutual fund industry may become more effective if advertising firms are aware of and apply the fundamental insights of Behavioural Finance theory. Besides more general variables of advertising effectiveness, private investors’ risk-return perceptions should be considered as special variables of advertising effects in the mutual fund industry, as such decision-relevant perceptions can be influenced by advertising stimuli. Private investors make use of judgmental heuristics during the processing of advertising stimuli, regardless of their expertise in investment decisions. Incompetent investors, however, make use of 56

Advertising research in the mutual fund industry heuristics to a larger extent, resulting in larger bias in the perception of risks and returns. These findings highlight the necessity of target group or market segment-specific advertising strategies in the investment industry, as differences in knowledge and experiences lead to different risk-return perceptions. Numerical values in print ads serve as anchor values and bias expected returns, even when there is no logical connection between anchor value and return estimation. Therefore, prominent numbers and figures in mutual fund ads have to be integrated very carefully, taking their potential biasing influence fully into account. Brand awareness and brand image play a central role in the processing of ads, as they are able to distort private investors’ risk perception at the product level. Investment risk is judged to be lower if the advertised mutual fund is offered by a highly reputable and wellknown investment company. As a consequence, investing in brand equity is important. Emotional states influence private investors’ risk perceptions. As emotional stimuli in the ad can lead to a more favourable, positive affective evaluation of the advertised mutual fund, emotional ads lead to a lower perception of investment risk compared to a merely informative advertising style. This indicates that emotional advertising is an effective tool, even in the abstract, rational, risk-return orientated investment industry.

Appendix Risk perception was measured by means of four items with regard to the different aspects of the construct, using five-point scales labelled “strongly disagree” to “strongly agree”: • • • •

“This mutual fund bears a high risk of losing money or of missing personal investment objectives” (downside risk) „I feel uncertain about investing in this mutual fund, as I feel uninformed and incompetent about it.” (ambiguity) „Investing in this fund also entails high chances to realise higher, above-average returns.” (upside risk) „Regarding this mutual fund, I reckon with high performance variations over time.” (volatility)

57

Klaus Peter Kaas • Jenny Jordan As all items related to one factor which we called “risk perception”, factor scores were calculated and used to form an aggregate perceived risk measure. Expected return as a direct numeric estimate was captured by the following question: “Please estimate the mutual fund’s average rate of return per year (time horizon: the next few years)”: on average _____ % per year.” Investors’ expertise in investment decisions and mutual funds, consisting of subjective knowledge and experiences (Alba & Hutchinson, 1987; MacInnis, Moorman & Jaworski, 1991), was measured by means of four items: 1. „How do you assess your knowledge about mutual funds?“ (five-point scale, labelled „I have a lot of knowledge“ to „I have very small knowledge” 2. „Please evaluate to what extent you have already gained experiences with mutual fund investments.“ (five-point scale, labelled “I am very experienced” to “I am very inexperienced”) 3. „I know a lot about investing money.“ (five-point scale, labelled „strongly agree“ to „strongly disagree“) 4. „I am very experienced in investing money“ (five-point scale, labelled „strongly agree“ to „strongly disagree“)

58

4

An Evaluation of Corporate Brand Character PERNILLE SCHNOOR

“Corporate branding is one of those things that everyone believes is important, yet there is very little consensus as to what it means. Words such as ‘values’, ‘image’ and ‘communication’ swirl around. It is undoubtedly related to all these things.” (Ind, 1997)

Abstract This article is concerned with the role of credibility in relation to corporate brands. By adapting the method of evaluating a person’s credibility, the article will present evidence in favour of developing a more comprehensive evaluation method. This method will show how corporate brands are perceived differently along five evaluative dimensions: success, credibility, exuberance, forcefulness and sincerity.

Background In recent years, a great deal of literature has focused on the effects and advantages of using a corporate branding strategy. Increasingly, companies seek to present themselves as corporate brands, following a corporate branding communication and management strategy. However, there is little empirical evidence of a corporate branding strategy’s positive impact on a company brand image. Focus is

59

Pernille Schnoor primarily on internal perspectives and how to manage corporate brands (Bengtsson, 2002). A key object in the brand management literature is how to evaluate brand equity. Brand equity can be seen as the set of assets linked to a brand’s name and symbol; assets that can add to the value provided by a product and/or a service (Aaker, 1996). Brand management literature is primarily concerned with product brands, and even though differences in product and corporate brands are pointed out, evaluation tools most often mix brands on a product and corporate level. Examples of this are Young & Rubicam’s Brand Asset Valuator (BAV) and Jennifer Aaker’s Brand Personality Scale (see, for instance, Keller, 2003). Corporate brand equity occurs when relevant constituents hold strong, favourable and unique associations about the corporate brand in memory (Keller; in Schultz et al., 2000). Hence, understanding how consumers and other stakeholders perceive corporate brands is an important research task. The literature provides no general definition of a corporate brand, but when looking at the term ‘corporate brand’ it might be useful to divide it into two parts, as done by Ind (1997): ‘Corporate’ implies organizations in their totality and ‘the idea of people coming together and working towards a common goal. ‘Brand’ implies something that is (…) “distinct from the simple idea of a product, in that there is a suggestion in the notion of a brand of values that go beyond mere functional performance” (Ind, 1997). A corporate brand can encompass a wider range of associations in the minds of consumers, and therefore it is distinct from a product brand (Keller, 2003). The literature reveals two overall understandings of corporate brands. One sees the company brand as the corporate brand and is typically called a company-as-brand-strategy or a monolithic corporate brand strategy. In this perspective, only one brand is communicated and the focus is very much on internal aspects, such as employees, corporate values, identity, etc (Chernatony, 2001; Ind, 1997). Examples would be The Body Shop and financial corporate brands such as mortgage-credit institutions. The other perspective sees the company brand as having an endorsement function behind other main brands and as the top-of-brand-hierarchy (Aaker &

60

An Evaluation of Corporate Brand Character Joachimsthaler, 2000; Keller, 2003). There are numerous examples of this strategy, such as Microsoft, Nestlé and Scandinavian Airlines (SAS). This latter understanding focuses very much on the market with the consumer as the key player. The two strategies are not mutually exclusive but have different focuses, which are relevant when dealing with different contexts, such as the specific product category and/or industry. Many researchers agree that rather than focusing on communicating individual product brands, corporations can benefit from focusing their efforts to communicate values embedded in the corporation as such (Ind, 1997; Thomson et al., 1999; Chernatony, 2001). This strategy is not only more economical – because communicating one brand is less expensive than communicating subbrands, brand extensions, etc. individually, but also a focus on the company values as such, rather than on the individual product benefits, can strengthen the perception and credibility of the company’s brands. The use of the company name reduces consumer insecurity and transfers credibility from the company to its products (Franzen & Bouwman, 2001; Aaker 1996). On the other hand, lack of corporate credibility can lead consumers to question the validity of claims by the company as such, and as a consequence consumers are less likely to buy its products (Franzen & Bouwman, 2001; Goldsmith, Lafferty & Newell, 2000). Thus, lack of corporate credibility can be crucial for a company and therefore credibility is a key notion in relation to discussing corporate branding.

Corporate brands and credibility According to Aaker (1996), credibility plays an important role in relation to defining and/or creating a brand identity. He mentions credibility as one of three elements in his brand identity system, the two others being core/extended identity and the value proposition. These are elements which will lead to a brand-customer relationship. Aaker also states that the credibility of a firm contributes to building brand equity for its products, both tangible goods and intangible services (Aaker, 1991). Whereas credibility of a product brand is based on more tangible features, such as product quality and technology, which are easier to 61

Pernille Schnoor evaluate for consumers – i.e. most products can be tasted, smelled or touched - credibility in corporate brands, such as service brands, is more complex, because it is based on intangible variables, such as people and systems (Clutterbuch et al. 1993; de Chernatony & SegalHorn, 2001). Corporate credibility relates to the reputation that a firm has achieved in the marketplace. In a marketing context, it refers to the extent to which consumers believe that a company is able to deliver products and services that satisfy customer needs and wants (Keller in Schultz, 2000). Newell and Goldsmith define it more broadly as the ‘extent to which consumers feel that the firm has the knowledge or ability to fulfil its claims and whether the firm can be trusted to tell the truth or not’ (Newell & Goldsmith, 2001). Unfortunately, the brand management literature does not provide much conceptual discussion of the credibility construct. Most research on the topic of credibility has been done in the field of communication and rhetoric.

Credibility and rhetoric As previously argued, consumers choose brands by corporations they trust. Credibility and trust are closely linked. According to Garver (1994), ‘when ethos disappears, so does trust.’ The discussion of ethos, or more popularly source credibility, dates back to Aristotle, who found that ethos is the most crucial factor in successful communication (McCroskey & Young, 1981). In Aristotle’s view, persuasion is created through the character (ethos) of the communicator, when s/he communicates in a way that establishes credibility (The Rhetoric of Aristotle, II-15). Aristotle found three important dimensions of ethos: intelligence or sound sense, moral character, and good will or benevolence (McCroskey & Young, 1981; Corbett, 1965). In a rhetorical perspective, the aim is therefore to create the impression by a discourse that one is a person of sound sense, high moral character and benevolence. According to Larson (1983), ethos consists of two elements, namely the persuader’s reputation and delivery: ‘Whether the persuaders are politicians, corporations, or organizations, they can all have an image or ethos, and it is based on their reputation as well as the delivery of the message.’ 62

An Evaluation of Corporate Brand Character Based on Aristotle’s thoughts on the importance of ethos, a massive scholarly interest has been shown in the topic of source credibility, which can be defined as ‘the attitude towards a source of communication held at a given time by a receiver’ (McCroskey & Young, 1981). For the last 50 years or so, communication scholars have studied the components of source credibility intensively. However, this research has mostly concentrated on persons. This research was initiated by Hovland et al. (1953) , summarized by, for instance, Andersen and Clevenger (1963), and followed up by many others, such as McCroskey (1966), Whitehead (1968), Tuppen (1974) and Hansen & Kock (2001). The number and type of credibility dimensions vary in the studies. Typically, two or three dimensions are identified for persons: for instance, competence and character (McCroskey & Young, 1981) or competence and trustworthiness (Bowers & Phillips, 1967).

The study The purpose of this study is to conceptualize and operationalize credibility in relation to the stakeholder perception of corporate brands. We hypothesize that the public image of spokespersons and of corporate brands can be evaluated by using the same method. Moreover, credibility may depend on different characteristics for different categories of corporate brands (Cronkhite & Liska, 1976). Hence, in trying to conceptualize a “corporate brand character”, we may expect a complex, multidimensional picture to emerge. Whereas studies of credibility, or other qualitative character dimensions of communicators, tend to produce factor analysis solutions with three or, at most, four dimensions, it is natural to expect an even larger number of factors when looking at “corporate character.” This is because corporations are not only communicators, however important that aspect might be, but also makers of physical products as well as sales organizations consisting of people. Thus, we hypothesize, the character of corporations is evaluated by the public along a broader range of dimensions.

63

Pernille Schnoor

Research questions To sum up: In the present study, we want to test the following: • Along which dimensions are corporate brands evaluated? • How is credibility perceived when evaluating corporate brands? • Are corporate brands perceived as more complex than persons are – so that more dimensions will appear? • Does it make sense to evaluate corporate brand credibility by using the same method as for persons?

Method In testing this, a large number of corporate brands should ideally be evaluated along a large number of scales. To do so would result in a questionnaire so extensive that it was feared it would probably influence the response rate and the quality of the responses. For this reason, a decision was taken to limit the study to 10 different corporate brands presented in pairs, which are related in their line of business but expected to have very different profiles. Following the tradition of source credibility research, a battery of relevant adjectives was constructed. The adjectives were adopted from previous studies on public spokespersons primarily done by Berlo, Lemert and Mertz (1969) and Hansen & Koch (2001). Berlo, Lemert and Mertz identified 128 pairs of polar adjectives that were frequently used to describe highly acceptable or unacceptable sources. This selection was done on the basis of interviews and literature reviews. The number of adjectives was reduced to 83 by asking faculty judges to group the scales on the basis of similarity in meaning. Berlo, Lemert and Mertz identified four factors, which they named: Safety, Qualification, Dynamism and Sociability. These factors relate to the credibility of persons. In order to adapt their study, which was concerned with the credibility of spokespersons, to this study of corporate brands, it was necessary to exclude adjectives that did not make sense. Adjectives such as stubborn, biased and intimate were therefore excluded from this study. In order to avoid too extensive a questionnaire, 38 adjectives were included in this study.

64

An Evaluation of Corporate Brand Character

Data analysis The study was carried out with the use of a questionnaire mailed to all new undergraduate students at Copenhagen Business School in the spring of 2001, and a total number of 169 useful responses were received (response rate 35%). Self-rating of the respondents’ awareness of the 10 corporate brands was also included in the questionnaire. Factor analysis was conducted for each of the 10 corporate brands. Factor analysis was chosen as the data analysis method because of its function of being able to identify underlying constructs in the data. In the factor model, there is a small set of independent variables, termed factors, which is hypothesized to explain or cause the dependant variable (here: source credibility). The aim is not so much to reduce the variables but to understand the interdependent variables in relation to the perception of corporate brands and credibility. Factor analysis also identifies variables that are redundant – measuring the same construct (Aaker, Kumer, Day, 2001). The factor analysis procedure employed in this study is principal component analysis. It appears that the solutions emerging here have significant similarity across individuals. With five factors, 30% of the total variance in the data is accounted for. The essence of this solution is shown in Table 4:1. Why five factors? There is no clear answer as to when to stop including factors. A rule of thumb is that all included factors (prior to rotation) must explain at least as much variance as an ‘average variable’. If a factor is meaningful and capable of representing one or more of the variables, it should absorb at least as much variance as an average original input variable. In this case, with as much as 38 variables, each of the five factors accounts for more than the average, i.e. here more than 3 %. In addition, Table 4:2 shows that after the fifth factor there is a ‘corner’ or a distinct break of the scree plot. Experimental evidence indicates that this point where the scree begins denotes the true number of factors. At the same time, the factor analysis shows that the content of the factors does not make sense when going beyond five factors.

65

Pernille Schnoor Table 4:1 – A five-dimensional solution successful intelligent strong powerful admirable visionary Self-confident purposeful ambitious knowledgeable professional efficient decisive serious organized responsible competent trustworthy correct sensible just active energetic colourful extrovert cheerful boastful arrogant aggressive superficial authoritative original capable of admitting mistakes warm honest sympathetic thoughtful open-minded

1 0.70 0.60 0.59 0.57 0.57 0.54 0.52 0.47 0.44 0.42 0.42 0.37 0.34 0.24 0.16 0.07 0.29 0.03 0.17 -0.09 -0.01 0.00 0.10 0.14 -0.01 0.07 0.17 0.18 0.12 -0.10 0.10 0.05 0.05 -0.03 -0.02 0.14 -0.01 0.06

2 0.02 0.06 0.06 0.05 -0.03 -0.03 0.11 0.21 -0.04 0.17 0.41 0.23 0.29 0.52 0.50 0.47 0.43 0.39 0.38 0.32 0.31 0.21 0.06 -0.10 0.13 -0.19 -0.10 -0.14 0.07 -0.15 0.24 0.11 -0.05 0.12 0.15 0.23 0.24 -0.17

3 0.02 0.03 -0.04 -0.13 0.12 0.22 -0.02 0.27 0.38 -0.04 0.00 0.16 0.06 0.04 0.15 -0.02 0.11 -0.07 -0.01 0.03 -0.01 0.61 0.60 0.55 0.50 0.49 0.07 -0.06 0.19 0.03 -0.06 0.15 0.04 0.10 -0.03 0.13 0.05 0.10

4 0.05 -0.03 0.15 0.34 -0.02 0.01 0.34 0.03 -0.02 0.07 0.08 -0.03 0.03 0.00 0.09 0.01 -0.07 -0.08 -0.10 -0.01 -0.07 0.03 0.01 -0.12 0.14 0.03 0.64 0.62 0.44 0.43 0.42 0.03 0.04 0.01 -0.06 -0.20 0.07 -0.12

5 0.00 0.05 0.03 -0.04 -0.04 0.15 0.07 -0.19 -0.10 0.13 0.08 0.04 -0.01 -0.10 -0.01 0.14 -0.07 0.17 0.23 0.04 0.15 0.01 0.00 0.18 0.21 0.26 -0.04 -0.03 -0.24 0.09 0.04 0.54 0.51 0.43 0.42 0.34 0.34 0.24

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.

One way to investigate how the 10 corporate brands are differently evaluated among these five dimensions is to calculate factor scores for the respondents on each brand, which is done in Table 4:3. This table shows how the different corporate brands are perceived in relation to the five factors. Whereas the insurance company Tryg is perceived as somewhat credible, this seems not to be the case for the insurance company Top Danmark.

66

An Evaluation of Corporate Brand Character Table 4:2 – Scree Plot

Scree Plot 6

5

4

3

Eigenvalue

2

1 0 1

5 3

9 7

13 11

17 15

21 19

25 23

29 27

33 31

37 35

Component Number

Microsoft scores very high on success and forcefulness, but very low on credibility, which is very different from the IT company Jubii, which is perceived as highly exuberant but not very credible.

Results A five-dimensional solution accounts for 30% of the total variance in the data. The variance might seem low, but it can be explained by the relatively high number of initial variables (38). The solution has significant similarity across individual corporate brands. We have labelled the five dimensions credible, successful, sincere, forceful and exuberant. This set, which is shown in Table 4:5, we suggest reflects how corporations are perceived as actors in a drama where many mutually independent aspects together shape the profile of any given character. Closer inspection reveals striking instances where two corporations, even in the same industry, have markedly different strengths and weaknesses (see Table 4:4). This means that if we want a useful representation of how corporations are perceived by the public, we need an analysis of corporate character along several

67

Pernille Schnoor Table 4:3 – Factor scores on the five dimensions for the 10 corporate brands Cumulative factor scores Factor 1 (success)

Factor 2 Factor 3 Factor 4 Factor 5 (credibility) (exuberance) (forcefulness) (sincerity)

DSB, railways SAS, airlines TRYG, insurance

-46.9289 50.77089 -57.1398

-7.96426 138.50256 50.30896

-1.5123 -6.3688 -35.70028

57.24812 -5.30379 8.41077

84.15601 -12.50966 10.34194

SILVAN, DIY centre

-54.8908 -40.4618

-16.36889 -47.68128

11.78852 72.66616

-34.51207 -19.2098

17.82281 -50.52329

-36.2668 36.90343 254.3846

18.74747 -114.75805 -41.11393

-48.30261 153.56481 -64.24009

-20.31004 -34.04615 118.3205

-41.41174 30.51399 -65.51323

-41.7527

49.78966

-45.0751

-33.00442

35.43083

-64.618

-29.46193

-36.81959

-37.59227

-8.30785

GO, airlines TOP DANMARK, insurance JUBII, IT MICROSOFT, IT POST DANMARK, postal service BYGGEKRAM, DIY centre

mutually independent dimensions. One or two dimensions will not suffice. Credibility, for example, may be equally high in two corporations, but for very different reasons.

Conclusion and implications Factor scores show that the 10 corporate brands are evaluated very differently along the five dimensions, indicating that this is a powerful and distinct instrument and that managers could learn much from using it. The study implies that this evaluation method can be used to describe a company’s character (- or image) relative to the other companies in the analysis. The method does not provide tools to describe a corporate brand character in relation to an ideal corporate brand character situation. Thus, it would be interesting to evaluate brands from the same industry in order to be able to compare the individual corporate brands to the perception of the industry as a whole. The present study has a number of limitations. Since we used existing items to measure credibility, it can be concluded that the quantitative analysis established the reliability and validity of the

68

An Evaluation of Corporate Brand Character concepts. However, it cannot be assured that we really discovered the existence of the concepts. The next step would be to apply the questionnaire to a second sample other than students. Table 4:4 – An example of the perception of three different corporate brands along the five dimensions Factor scores on the five dimensions for three different brands 3 0 0 ,0

D SB , railw ays

2 5 0 ,0

S AS, airlines

2 0 0 ,0

MIC R OSOFT, IT

1 5 0 ,0

1 0 0 ,0

5 0 ,0

0 ,0

- 5 0 ,0

- 1 0 0 ,0

Succes

C redibility

Exuberance

Forcefulness

Sincerity

Table 4:5 – The five dimensions of perceived corporate brand character Success

Credibility

Serious, organized, responsible, competent, trustworthy, correct

Exuberance

Active, energetic, colourful, extrovert, cheerful

Corporate Brand

Successful, intelligent, strong, powerful, admirable, visionary, selfconfident, purposeful, ambitious, knowledgeable

Character

Forcefulness

Sincerity

Boastful, arrogant, superficial, authoritative

Original, capable of mistakes, warm, honest

aggressive,

admitting

69

5

Barriers to e-Commerce GORM KUNØE

The barriers to e-commerce that have an impact on young people’s buying behavior on the World Wide Web (WWW) are investigated in an exploratory research design. A factor analysis is carried out to examine propositions on five barriers to e-commerce. The analysis placed the factors in the following range of importance to buyers on the WWW: (1) patience, a new term to marketing research, (2) trust, (3) motivation, (4) Internet maturity and (5) the effect of traditional shopping experience. The findings allow for a theoretical explanation on how these barriers are formed and how theories of relationship management and one-to-one communications can be used to reduce these barriers. The results indicate that companies should focus on the barriers of e-commerce as fundamental preconditions for their ecommerce strategy. The managerial implications from the barriers are extensive, serious and affect any e-commerce vendor. E-commerce has already changed the world of buying and selling (Negroponte, 1995; McKenney, 1995). The World Wide Web (WWW) and @mail open radically new opportunities for consumers and marketers to communicate in dialogue and to buy and sell via the screen. Yet users of the Internet do not shop indiscriminately or leave private information to the vendors at will. There are profound barriers to e-commerce of significant importance to the vendors and to governments who have a keen interest in promoting e-commerce.

70

Barriers to e-Commerce The e-commerce web site tries to make the virtual shop look like the shop with which we are familiar. The buyer fills a “basket”, the tone of voice is polite and much is done to strengthen familiarity on the web site, with the traditional shopping environment as a model. Visible attempts are made to build relations on the WWW by collecting lifestyle and other kinds of personal information not directly connected to the purchase, in order to get closer to the customer. As a concept the web store seems to be a flawless place to shop. Recent research has been devoted to studying how the optimal pages should look and function (Nielsen, 2000), and the web site visitor’s clicking behaviour (Murphy, 1999) is engaging Web analysts all over the world. However, the users are not as frantic as the vendors, and ecommerce companies go broke every week. Research shows (MMI, 2001) that 74% of the Norwegian respondents were deeply concerned over the possibilities of handing over their credit card number to any vendor on the Internet. As the chance of having one’s credit card misused is fractional, there might be a more complex resentment or uncertainty towards e-commerce, which has passed unnoticed by the vendors. Consequently, we set out to explore the perceived barriers or barrier system to e-commerce. We felt that a series of barriers might be inherent in the mind of the shopper, who has the good old physical shop in mind when shopping on the WWW. By revealing the barriers to e-commerce, we also gain a new perspective on our relationship building and the handling of data in a traditional shopping environment. All over the world, considerable data analysis is systematically carried out to track where and when potential shoppers leave an e-commerce web site. In an example from ACNIELSEN NetRatings1 (March 2001), characterised as being a typical example, an American pet food store on the Internet experienced that 6% of the visitors on its web site finalised the purchase. 80% left the site before putting anything in the “basket”. Of the remaining 20% who placed something in the “basket”, 10% subsequently left the site. At this stage, 10% of the users still remained on the web site, who should thus be a good prospect for a purchase. This group proceeded to the payment site, of whom 4% left without paying. The remaining 6 % are buyers. The example illustrates the 1

We are thankful to ACNielsen NetRating, Norway for sharing this information with us.

71

Gorm Kunøe problem: Is this the best result the pet shop could expect? Or is it a poor outcome compared to what is happening when people visit the real store? Dog lovers seldom take a stroll through the pet food store just to have a look. They either stay outside or go inside to buy something.

Literature Review A well-functioning relationship between seller and buyer is acknowledged as a cornerstone of marketing, and something most marketers of today want to achieve. The development of the WWW has accelerated the possibilities of being able to approach individuals by the million and still be personal and relevant. It is possible to copy the best of relationship building from real shops onto the web shop. The roots of what is sought for on the WWW and on print as One-to-One Marketing (Peppers and Rogers, 1993) are far-reaching, and should be understood by e-commerce vendors. With the paradigm shift from a transactional orientation to a relations orientation (McKenna, 1991; Gummesson, 1995), another way of influencing the customers through dialogue was presented as One-to-One communication by (Peppers and Rogers, 1993) and Dialogue Marketing by (Kunøe, 1989) using traditional methods from communication in industrial marketing in the business-toconsumer market. One of the common features of these theories is the more or less explicit “revolt” against mass communication in any media, including the WWW, and its lack of ability to establish and effectively maintain relations. Examples of different contributions on more or less the same issue include: issues on relationship selling by (Evans, 1963); relationship marketing by (Håkonsson, Johanson and Wootz, 1976), (Christopher, Payne and Ballantyne, 1991) and (Gummesson, 1995); direct marketing by (Nash, 1986), (Stone, 1988), and (Roberts and Berger, 1989); interactive marketing by (Blattberg and Deighton, 1991) in “Cyperspace”, as well by Roehm and Haugtvedt; and database marketing by (Shaw and Stone, 1988). The area of dialogue marketing is represented by (Kunøe, 1989; 1998) and by (Pine, 1993). Advocates of individualised marketing and marketing communication are (Schultz, Tannenbaum and Lauterborn, 1993) and (Hutton, 1996), being examples of closely related ideas. Our research into barriers to ecommerce set out to reveal what antecedents this relationship building

72

Barriers to e-Commerce will have to take into account. (Van den Poel and Leunis, 1999) have studied the WWW as a channel of distribution and show that heavy users of the Internet evaluate the Internet more favourably than light users do. They also researched the risk perception and found that users have a clear notion of the risks involved in e-commerce. We present and discuss the antecedents to the attitudes towards ecommerce by frequent and less frequent users of e-commerce. In addition, we discuss the individual barriers, present the results, and explore the implications of the results for further research and managerial practice.

The Research Agenda In the context of this study, a barrier is defined as a relatively wellconsidered negative reaction of complex nature to an e-commerce offer. Using this definition, the barriers we have researched are more than temporary negative reactions to a single purchase offer. Consequently, our findings have to be taken more seriously if they are to be used as input for an e-commerce strategy than if our aim was to research minor adjustments to those web sites which cause irritation and defection. One of our research issues is the degree of the uniqueness of the WWW and e-commerce pages as commercial tools. Does the WWW constitute a new media with characteristics that make the use of traditional communication theory and models impossible? Are we in need of a particularly new approach to theories and methodologies if the WWW itself is a paradigm shift? Or is in fact ecommerce just another ordinary marketing tool, subject to the same kind of profound understanding of how to treat consumers? Our main findings fall within five barriers and are identified as patience, trust, motivation, Internet maturity and the effect of the shopping experience. In Table 5:1, we present the questions and the results of the factor analysis.

Patience The term “patience” is a seldom discussed and rarely used variable in marketing research. It is a term focused on in the social cognition literature (.e.g Augoustinos and Walker, 1995) in connection with

73

Gorm Kunøe understanding the attitude-behaviour link. “Patience” as a research term should be more focused as an explanatory variable, because Internet users have their patience tried by marketers on the WWW as often as the users log on to the marketer’s web site. A negative experience with a web site does not build a positive attitude towards the brand. The feelings connected to time spent in front of a computer screen can be divided into two main areas of interest to our research. One area is the positive or negative feelings connected to waiting time spent in interaction with the application. The other area is the duration of forced idleness, where one waits for something to happen on the screen. Some of this idleness is self-inflicted by the user or forced upon the user by the application. Whoever is the cause of the wasting of the user’s time, the feeling of wasting one’s time is a barrier to ecommerce. The term “patience” in our research is connected to the subjective reflections and reactions made by the potential buyer on the following variables: • Experience: How much time does the individual expect to use on this kind of operation? • Functionality seen from the buyer: The individual understanding of the potential buyer on how to navigate the web site should meet the competence of the individual. If it does not, the individual does not always blame him/herself for the lack of skills. Whether or not “functionality” can be seen as a barrier to e-commerce is a judgement by the potential buyer on the WWW. • Necessity: The importance of the purchase has an impact on the amount of patience the potential buyer feels can be set aside for the operation. The urgency can in a “patience context” lead to an extension of patience or to the opposite, namely impatience. In our questionnaire, “experience” and “functionality” were phrased as examples of scaling of “time spent” in the opening of an e-commerce page and the navigation. We asked how impatient users felt and how this affectedtheir behaviour and attitude towards e-commerce. There is a significant correlation between these two questionnaire items: “How often are you on the Internet?” and “If the web site is difficult to navigate, I get easily annoyed.” The more often the respondents are on the Internet, the easier they become irritated when they have difficulty navigating the site. One could assume that frequent use would make 74

Barriers to e-Commerce the customer more indulgent. However, that does not seem to be the case. The wording of our questions for the factor analysis can be found in Table 5:1. P1: The factor “patience” is related to the psychological state into which the user of a web site is moving when the user has spent time waiting for a web page to open and when navigating the web site. This mood raises a barrier to e-commerce.

Trust Trust is considered a major variable in the understanding (Morgan and Hunt, 1994) of how to establish, develop and maintain successful relational exchanges in Business-to-Business. Within marketing research, trust has been subject to some research from various focus points. (Dwyer, Schurr and Oh, 1987) focused on commitment, as did (Morgan and Hunt, 1994). In our e-commerce context, trust should be viewed from the supplier’s as well as from the Internet user’s point of view, an interesting area of research which we have as yet had no possibility of pursuing. Our trust variable solely rests on the opinion of the user of e-commerce. (Doney and Cannon, 1997), researchers in social psychology, define trust as the perceived credibility and benevolence of a target of trust. The objective credibility is the expectance within the mind of the buyer that the other partner’s word or written statement can be relied upon. The benevolence element in the definition stands for the extent to which a partner is genuinely interested in the other partner’s welfare and joint business prosperity. Trust is thus a matter of not being misused either in economic matters or in other matters where one has placed oneself open to distrust. The user’s perception of the seller’s attitude to the possible misuse of information given in confidence, such as credit card numbers and personal information, was tested in the questionnaire. Trust consists of the reciprocity in the dyad, where the buyer delivers economic and personal data which enables the seller to withdraw money from the buyer’s bank account and continue a marketing process. An intriguing difference between the trust antecedents discussed by (Doney and Cannon, 1997) is the intuitive form trust has with the buyers on the

75

Gorm Kunøe WWW. Especially newcomers to e-commerce do not have an established relationship with their e-commerce supplier, except with their bank, and therefore are reluctant to trust what goes on in ecommerce as such. The lack of a common relational history in the buyer-seller dyad darkens the potential buyer’s attitude towards ecommerce as a trustworthy way of buying products. P2: Trust is a relationship building variable in a fair and reciprocal context of e-commerce trade, where valuable information and money is given in return for goods and services. Lack of trust contributes to the building of barriers towards e-commerce in the mind of the buyer .

Motivation Motivation in an e-commerce context is the inclination of acting according to one’s attitude towards the seller. In our context, to “act” means to buy something on the Internet or to register information or not. In such a concrete setting, it is possible to pose questions which are relevant to the respondent. We asked three questions: To what extent was their buying behaviour influenced by (1) the lack of visual experience with the product, (2) the intensity of the wish to try the product, and to what extent (3) did a general uneasiness towards the quality of the product influence their behaviour on the screen. P3: The motivational factor consists of the extent of the perceived lack of visual experience with a product. The intensity of the wish to try the product, and the extent a general uneasiness towards the quality of the product influenced the potential buyer’s behaviour on the screen, contributes to raising a barriertowards e-commerce.

Internet Maturity The use of the Internet as a buying tool is new to many people. Norwegian research (SIFO 1999) shows that the three most preferred products purchased on the WWW were CDs, technology and data equipment. All products can be described as being of relatively inferior

76

Barriers to e-Commerce value, so if one was cheated, it would not matter much. The research shows that the interest in e-commerce is keen. However, the respondents distrust the vendor for having motives one would encounter among persons one did not know. (Van den Poel and Leunis, 1999) show that use itself of the Internet will relieve some of the perceived risk over time. Consequently, a variable of major importance to the use of e-commerce could be practice and the fact that one becomes more and more accustomed to the WWW through use. The outcome of trial and error over time construct the individual’s learning curve, a curve over which the vendors on the WWW have possibilities to gain some control. This is what we refer to as “Internet maturity”. The maturity parameter has a dynamic effect on the learning curve, which constantly receives input from various sources of marketing and the use of the Internet. In order to cope with the challenge of identifying a rather intangible parameter, we asked our respondents about their feelings related to the use of e-commerce web sites. Three questions were asked about the buyer’s feelings and experiences from e-commerce. The first question (1) asks how irritated the respondent becomes if the site is difficult to navigate. The second question (2) asks to what degree one defects from the site when the site seems difficult to navigate. Finally (3), we inquired about the consequence of bad experience with a web site difficult to navigate. We asked to what extent they hesitated when the experience was negative. P4: Three terms can contain the phrase “Internet maturity.” It can be expressed at the level of irritation when web sites are difficult to navigate. Furthermore, the extent to how quickly one leaves the site when it is found to be difficult to navigate and the extent the bad experience with a certain web site influences the inclination to log on to the vendor’s web site contributes to the building of barriers to e-commerce in the mind of the buyer.

Traditional Shopping Experience The shopping experience from real shops is one of the foremost comparisons buyers make when clicking onto a web site. We have sought to gather all the impressions from real shopping by forcing the

77

Gorm Kunøe respondent into a relational mode by recalling the shopping experience. The seller, being aware of that, tries to build a shopping environment similar to the one encountered in real shops to help the recollection of a positive, safe shopping environment. As a screen shop experience is rather different from the experience of a real shop, there could well be a barrier towards this new buying media. Other relevant comparisons on behalf of the potential buyer are experiences with other e-commerce web sites, a point we shall leave to further research. We constructed five variables to cover some of the contents in the phrase shopping experience. The first question asked fathoms the respondent’s attitude towards Internet shopping as such. The second and third question, respectively, concerned to what degree the wish for shopping in a real shop discouraged Internet shopping, and to what degree abstention from e-commerce is caused by general negativity towards mail order, which is a close comparison to e-commerce. We asked if the respondent did not buy on the Internet because of general uncertainty of how e-commerce functions. Finally, we asked about the importance of the security regarding personal information delivered. The following proposition was formed: P5: The positive relational feeling from shopping in ordinary shops is a barrier to e-commerce and can be explained by five variables. The barrier can be identified by an expression that includes a person’s general attitude towards e-commerce. The extent of preference towards ordinary shops, the degree of a positive attitude towards mail order as such, the degree of uncertainty towards how e-commerce functions, and the extent to which the respondent wishes to be free of unwanted mail after having used e-commerce, contribute to the building of barriers to e-commerce in the mind of the buyer.

Method The research method is an exploratory, quantitative research design based on five propositions. Data were collected by a questionnaire distributed to 199 students aged 19-27 years. All studied at The Norwegian School of Marketing in Oslo. Three classes were selected for the questionnaire: One part-time class with older students all

78

Barriers to e-Commerce having full-time jobs, a standard class of full-time students specialising in economics, and one class from the School of Marketing. The two latter classes were approx. five years younger than the part-time class. We anticipated that there would be interesting variations between the classes, but no variations were found. The questionnaire contained 45 questions which measured on a Likert scale from 1-7. Included in the number are six questions regarding the user’s Internet access, their frequency on the Internet, as well as their gender, age and income. What we present here is the factor analysis being the quintessence of the research. Various other results of minor interest to our research are omitted.

Results Table 5:1 shows the results of the factor analysis from questions 3, 5 and 6 of the questionnaire. The eigenvalue should be larger or similar to 1 (Kaisers’ rule). When we set this as a precondition, we obtain these results in our five-factor structure. We conducted a factor analysis on the five parameters: patience, trust, motivation, Internet maturity and the effect of shopping experience. These barriers are thought to be crucial to the way people react to e-commerce as such. The results are presented in two ways: Statistics and cross tabulations in the text and the factor analysis in the form of Table 5:1. The structure we obtain is statistically significant, which should meet a demand of significance >0,05. A general rule is that the larger the significance, the larger the probability that the structure is correct. The five-factor structure explains 58% of the total variance. The factor structure is based on a Maximum Likelihood Extraction. Goodness-of-fit Test Chi-Square df 79.944 73

Sig. .270

Patience is the highest ranked factor. 84% (valid percentage) of the respondents who are on the Internet daily do not want to wait for a web site to open. This is 10% more than those respondents who are on the Internet once a week. 78% of the daily Internet users and 72% of the

79

Gorm Kunøe Table 5:1 The results of the factor analysis Factor 1 Patience 3.1) I think it is positive that one can purchase goods and services on the Internet 3.2) I think that the seller on e-commerce web sites uses my information for other purposes than I expect 3.3) It is safe to use one’s credit card when buying on the Internet 5.1) I have not been buying on the Internet because I distrust the security 5.2) I have not bought anything on the Internet because there is a risk of my credit card being misused 5.3) I have not traded on the Internet because I will not give away my credit card number 5.4) I have not traded on the Internet because I want to see the products before I buy 5.5) I have not traded on the Internet because I want to try the products before I buy 5.6) I have not traded on the Internet because I am not sure about the quality of the products 5.7) I have not traded on the Internet because I prefer to buy in a real shop 5.8) I have not traded on the Internet because I do not like to buy goods on mail order 5.9) I have not traded on the Internet because I am insecure about how the Internet functions 6.1) If the web site is difficult to navigate when I am going to buy, I get easily irritated 6.2) If the web site is difficult to navigate when I am buying on the Internet, I easily defect and leave the site 6.3) If my experience with the navigation of a special web site is negative, I hesitate to log on again 6.4) I think it is o.k. that it takes some time to .760 open the site when I am going to buy something on the Web 6.5) I think it is o.k. that it takes some time to .999 find my way into the set up when I am trading on the Internet 6.6) It is important to me that the seller guarantees that I do not get any unwanted mail

Factor 2 Trust

Factor 3 Motivation

Factor 4 Internet maturity

Factor 5 Shopping experience -.420

.356

-.615 .858 .927

.907 .770 .817 .634 .525 .361 .385 .796 .681

.575

-.241

weekly users disliked having to spend time navigating their way through some set-up on the page. The large majority of Internet users do not want to wait. As the patience factor is as important as it seems, the functionality of the web site should be examined very closely in

80

Barriers to e-Commerce order to meet customers’ demand for easy and logical access. The proposition of “patience” is verified. Trust is the second factor and has five variables connected, which is three more factors than the two contained in the factor “patience”. The cross tabulation between the frequency of use (daily, weekly, more seldom) and the belief to what degree the buyer expects the vendor to misuse information from e-commerce is interesting. There are no significant differences depending on how often one buys on the web. 53% of the three highest grades on the Likert scale state that they think information is misused. One would think that frequent use of the Internet would reduce our proposition, but this is not the case. This could mean that the distrust is relatively profound and must be seriously addressed by the vendors. There is no significant correlation between the feeling of how safe it is to use one’s credit card in an ecommerce transaction and how frequent the respondents trade on the Internet. Measured on the three lowest ranks on the Likert scale, 53% do not feel safe when using their credit cards in an e-commerce transaction. An interesting observation is that there is a gender difference in the attitude towards the use of credit cards for Internet purchases. 45% of those who felt that using credit cards is safe are men, while 24% of the same sample are women. The proposition is valid. There is a relatively huge barrier of distrust towards e-commerce which blocks a normalisation of trade on the Internet. Vendors should concentrate on actions and communication about security and seek to further legislation that can punish abuse. The third ranking of our factor analysis is the motivation to buy on the Internet. Our findings show that there are at least three motivators to e-commerce. The buyers want to see the products, they want to test and try, and they are not sure about the quality they get. 34% of the respondents did not answer questions related to the motivation factor. We are here at a crucial point in relation to e-commerce: the difference between selling a virtual product and the “real thing” you can test and try. We asked how important it was for the respondents that the web site was marked with a security guarantee and a money-back guarantee. 90% answered that they were interested in these guarantees. However obvious this question is, such guarantees must be taken into account, as they are keys to one of the barriers.

81

Gorm Kunøe Great expectations were connected to results connected to the “Internet maturity” factor. Especially in times where e-commerce is taking its first staggering steps, it is interesting to research into the variables that constitute “maturity” on the Internet. The three variables analysed show that our proposition is valid. Irritation caused by low perceived functionality on a web site will grow when the maturity variable increases. What we today excuse as teething problems to e-commerce web sites will not be tolerated in the future. And functionality will hopefully be improved on a one-to-one basis.

Discussion Summary of Results Overall, the results provide strong support for our five propositions and help clarify the roles of patience, trust, motivation, Internet maturity and the effect of traditional shopping experience on e-commerce. Our exploratory study indicates that (1) there are perceived barriers to ecommerce among a group of people who are in the main segment for the vendors on the WWW. These barriers are psychological and hinder the development of e-commerce. In addition, they are not connected to any special brand or organisation selling on the WWW. Our results show that the relationship between the user and the vendor on the WWW can be compared to the absence of a one-to-one relation or what one experiences in a mass marketing context. The hybrid of intimacy many e-commerce vendors of today are exposed to has little to do with one-to-one communication, dialogue marketing or relevant interactive marketing. Our findings show: (1) that instead of a strong inclination to buy on the WWW, young people feel very uneasy and refrain from buying; (2) that the “tallest” barrier found in our study is patience, a new term to marketing research. The users of the Internet are reluctant to try and retry a web site - as long as they think it does not meet with their expectations of navigational effectiveness and efficiency. (3), that when we view the four remaining barriers: trust, motivation, Internet maturity and the effect of traditional shopping experience, they are part of relationship management in a direct marketing context; and(4) that shopping on the WWW is new to the world, but e-commerce does not overrule important fundamentals of ordinary trade.

82

Barriers to e-Commerce

Managerial Implications To the vendors on the Internet, the barriers explored in our study are lethal threats to their business. Patience is the most important barrier found in our study. Potential Internet buyers react when the vendor wastes their time, and now we know that they punish him by not logging on again. This must be dealt with by the vendors, and should be relatively easy to target. Increased effectiveness and better functionality, based on customer orientation and not on technical limitation, must be the goal. When a marketing organisation has a profound trust problem, as revealed in our study, the organisation must take action to remedy this situation. If the vendor wants to motivate its users to shop, the vendor must present the product to the potential buyer in a traditional way. This could be seen as a shortcoming of the virtual community where the consumer still wants to try the shoes or the shirt before making a purchase. The motivational factor is connected to the factor that had the lowest ranking in the factor analysis, traditional shopping experience. Together, the two results show that potential shoppers on the Internet want to test before buying. It is then up to the vendor on the Internet to arrange for this possibility, which among others Amazon.com has planned to do by purchasing a building for a traditional bookstore.

Limitations and Further Research The relationship marketing and dialogue marketing literature deal with the five factors analysed in our study. At the start of the second millennium, e-commerce vendors and e-commerce research seems much occupied with the right way of measuring user behaviour on web sites; a stronger focus on why people do not buy in the first place, and if they buy why they restrict their purchases to low-cost items, such as books and CDs. Further research on how the five factors interact is planned. We still do not know what happens when one factor is fulfilled, and we do not know if the five factors are interlocked, although the respondents have ranked their answers. The result of this study form part of our research programme, “Dialogue on the Internet”.

83

6

Advertising research: a case study of Lithuanian breweries LAIMONA SLIBURYTE & REGINA VIRVILAITE

Introduction To act in the market successfully one has to know what motivates a consumer to buy. A number of questions have to be answered: what needs of the consumer the product or service will satisfy? How to meet his needs in a differential way and better than the competitors do? How to reach the customers? Various research - of product, market, customer, promotion strategy and effectiveness - help to find answers to these and many other questions. The latter ones can help to evaluate the effectiveness of advertising before using the means of media, during the advertising campaign or after it. Is it necessary to analyse the effectiveness of advertising? Usually, certain goals are set for the employees to achieve. Evaluation of their work is based on employees’ ability to achieve the goals. Advertising is not an exception. It is important to estimate whether the communication programme functions well and to assess its effectiveness by research. Advertising is a connective link between producer and consumer. It is an especially effective means of advertising retaining existing consumers and attracting new ones. At the same time, advertising becomes one of the most effective means of competition. There is a constant search for ways to measure the effectiveness of advertising. By the application of various methods, both clients and agencies try to find out whether their messages are presented

84

Advertising research: a case study of Lithuanian breweries effectively. Some companies have chosen a research methodology most appropriate to them and also useful to the companies conducting the research. Research conducted before advertising and after it allows for estimating its effectiveness in perspective and avoiding “costly” mistakes in the future: wrong advertising can do much harm for the image of a company and reduce demand for its production. During the last decade, fundamental changes have taken place in the Lithuanian beer market. In order to acquire or maintain a competitive advantage, more and more often, local breweries use foreign investments. Such well-known Scandinavian breweries as Carlsberg, BBH, Olvi, Orkla, Danish Brewery Group A/S and others became the investors. The Lithuanian beer market is rather special, as it developed much more rapidly than other branches of the economy. Foreign investors not only invested in new technologies, but also generated very active competition. Today with intense competition in the Lithuanian beer market, advertising becomes one of the most important means to inform, present and generate interest, as well as convince the customer. Advertising helps the customer to choose a product, encourages the purchasing of it and acts purposefully.

Background Lithuanian beer market analysis The Lithuanian beer market has grown significantly within the last few years and today’s per capita consumption is 59 litres. The increased beer consumption is mainly due to the economic growth in Lithuania and generally improved quality of production. Lithuania is still a growing market and an annual 5-7% increase in beer consumption is expected during the next five years. Growth of the sector was mainly due to large foreign investment, which facilitated the renewal of the technological equipment in the breweries, improved the quality of the beer and increased sales. Foreign investment in food and beverage is the largest in Lithuania. In 2001, Baltic Beverages Holding (a Pripps-Hartwall company) invested USD56.3m in the Kalnapilis and Utenos breweries. Another worldfamous brewery, Carlsberg, invested USD47.1m in Svyturys brewery.

85

Laimona Sliburyte • Regina Virvilaite In 1998 alone, general foreign investment in the country reached LTL764.8 million or 36.31%. Today there are about 80 beer production licensees in Lithuania. 5% of the market belongs to the small-size breweries. Ten breweries dominate the Lithuanian beer market; their production capacity is over 20 million dekalitres. According to data from the Department of Statistics, 18.343 million dekalitres (dal) of beer was produced in 1999, which is 17.7% more than in 1998. According to data from the Lithuanian Association of Brewers, the ten largest Lithuanian breweries produced 17.681million dekalitres of beer in 1999, which is 20.3% more that in 1998. Companies working in the Lithuanian beer market can be divided into two large groups: producers and importers (see Table 6:1). Table 6:1 Producer companies and importer companies in the Lithuanian beer market. Importers

Producers of beer Main brewers AB Kalnapilis

Beer brewers in rural areas Rinkuškiai

AB Svyturys AB Utenos alus AB Ragutis AB Vilniaus Tauras AB Gubernija AB Žalsvytis

Senas Malūnas Panoras Vilkmergė Dalitas Jūsų geradarys IĮ Mintis

AB Mažeikių Lokys UAB Biržų alus UAB Daiga

Stumbras Žaldokas Gesevičiaus IĮ

Amstel Bierbrouwerij Zoeterwoude Anheuser-Busch Faxe bryggeri A.S. Mast-Jagermeister Sinebrychoff Holsten Ceres Breweries Ltd. Aarhus Zipfer Aldaris Tuborg Calsberg

Our research shows that three large breweries, major competitors, produce the largest amount of beer in the Lithuanian beer market, i.e. Utenos alus (until 1998 called Utenos gėrimai), Kalnapilis and Svyturys. Several middle-scale breweries are active in the market as well. These also compete among themselves and plan to get a bigger market share in the future. They are Vilniaus Tauras, Ragutis, Gubernija, and small-scale breweries, who mainly sell their production locally. The above-mentioned three largest breweries have the best technological and production potential. Beer brands produced at these

86

Advertising research: a case study of Lithuanian breweries companies are being sold in the whole of Lithuania and, as research shows, are popular among beer consumers. Such a distribution of breweries acting in the market necessitates a further analysis. Analysis of the Lithuanian beer production market focuses on the activity of major companies (large and middle-size group breweries) in the market. We can make an assumption that small-scale companies and import, at the time of analysis, cannot make a significant impact on the competitive environment in the beer market. The following represents the structure of the Lithuanian beer market from 1997-2001. Figure 6:1 Structure of the Lithuanian beer market in 1997 30

27

23

Market share, %

25

21

20

11

15

7

10

3

5 0

Utenos alus Kalnapilis

Švyturys

Ragutis

Vilniaus Tauras

Gubernija

In 1997, the ten largest Lithuanian breweries produced 136.1mion litres of beer. That was 20.5% more than in 1996. Kalnapilis became the market leader by producing and selling 36.4 million litres of beer and took 27% of the market share. The Panevezys region competed with their largest brewery Utenos alus. This was due to the decreased production and sales of the latter to 31.78 million litres of beer. In 1997, the difference between the largest breweries and other breweries was quite significant. In 1997, the market share of Klaipeda’s Svyturys, compared with 1996, increased from 15.5% to 21%. In the same year, Kalnapilis, Utenos alus, Svyturys and Ragutis produced 110.85 million litres of beer and all these companies together had 82% of the market. The structure of the Lithuanian beer market was a tight 87

Laimona Sliburyte • Regina Virvilaite oligopoly with one dominant company Kalnapilis. In 1997, the company Utenos alus lost its leading position in the beer market; its market share compared to 1996, decreased from 35.1% to 23%. The same year, most companies, not only the largest, but also smaller ones, such as Vilniaus Tauras and Ragutis redesigned their production and thus improved the quality of their beer. The modernization of production significantly affected the marketing strategies applied by the breweries. Figure 6:2 Structure of the Lithuanian beer market in 1998.

24.4

Market share, %

25

19

20 15

12.7

9

7.4

10

4.8

5 0

Utenos alus Kalnapilis

Švyturys

Ragutis

Vilniaus Tauras

Gubernija

Analysis of the structure of the Lithuanian beer market in 1998 (see Figure 6:2) supports the conclusion that the structure remained the same: a tight oligopoly with strong domination of one company. Although the Kalnapilis market share in 1998, compared to 1997, decreased from 27% to 24.4%, the company remained in a leading position in the beer market. Market shares of other large breweries decreased as well. Svyturys market share in 1997, compared with 1998, decreased from 21% to 19%, and the Utenos alus market share was only 12.7%. Such a significant decline of the market share of the brewery was influenced by the reconstruction of the company, which was aiming at increasing its production capacity. The company experienced difficulties working to full capacity. The market shares of middle-size breweries Ragutis, Vilniaus Tauras and Gubernija increased slightly. 88

Advertising research: a case study of Lithuanian breweries Figure 6:3 Structure of the Lithuanian beer market in 1999.

Market share, %

25

20.8

20

19.5

20 15

9.2

6.4

10

5.6

5 0

Utenos alus

Kalnapilis

Švyturys

Ragutis

Vilniaus Tauras

Gubernija

Analysis of the Lithuanian beer market structure for 1999 shows (see Figure 6:3) its change compared to the situation in the market in 1997 and 1998. Figure 6:4 Structure of the Lithuanian beer market in 2000. 28.1

30

Market share, %

25

22.1

19.9

20

11.7

15

8.1

10

5.2

5 0 Utenos alus Kalnapilis

Švyturys

Ragutis

Vilniaus Tauras

Gubernija

There was no one dominating company, as the market shares of the largest breweries were very similar: Utenos alus had 20.8%, Kalnapilis – 20.0%, Svyturys – 19.5%. Analysis of market shares reveals that the Utenos alus market share had significantly increased, as did the Svyturys share, while the Kalnapilis market share had decreased. We can conclude that the changes mentioned were influenced by the altered competitive behaviour of Utenos alus, namely by increased 89

Laimona Sliburyte • Regina Virvilaite production capacity after the reconstruction and the successfully implemented strategy of market differentiation. Small-scale breweries Vilniaus Tauras and Gubernija increased their market shares, while the market share for Ragutis decreased to 6.4%. Analysis of sales of the main Lithuanian beer producers in the year 2000 supports the conclusion that Svyturys became the market leader moving from third position in 1999 to the first in 2000 and taking 28.1% of the market. Shares of the other two largest beer producers changed insignificantly: Utenos alus reached 22.1% in 2000, and the Kalnapilis share dropped to 19.9%. Although Kalnapilis changed its marketing strategy and introduced new brands of beer, nevertheless strong competition in 2000 and weak efforts to establish the Kalnapilis image in the market influenced the decrease of its market share. The structure of the market can be described as a tight oligopoly with one dominant company: Svyturys.

Market share, %

Figure 6:5 Structure of the Lithuanian beer market in 2001

30 25 20 15 10 5 0

29.4

18.4

17.5 12.9

7.5

Utenos Kalnapilis Švyturys alus

Ragutis

9.2

Vilniaus Gubernija Tauras

In 2000, there were no significant changes within the group of middlesize beer producers: Ragutis and Vilniaus Tauras increased their market shares up to 8.1% and 11.7% respectively, while the share of “Gubernija dropped to 5.2%. In 2001, Svyturys remained the market leader with a slightly increased market share of 29.4%. The market structure remained unchanged – an oligopoly with one dominant company. The market share of Utenos alus dropped to 18.4%; the Kalnapilis share dropped to 17.5%. Within the group of middle-size breweries, the market share of Gubernija changed by increasing to 9.2%. The shares of the other two 90

Advertising research: a case study of Lithuanian breweries companies increased insignificantly: The Ragutis share dropped down to 7.5%, and the Vilniaus Tauras share increased to 12.9%. Our analysis shows that competition between Lithuanian beer producers is very high. The main strategic investors in the Lithuanian breweries are Baltic Beverages Holding, Carlsberg, Olvi, Faxe Bryg Holding and others. The joint Finnish-Swedish company, Baltic Beverages Holding, invested in the Kalnapilis and Utenos alus breweries. In June 1999, the Danish Company, Carlsberg A/S, which owns the Estonian brewery, Koff, acquired Svyturys of Klaipeda. In September 1999, the third largest Finnish brewery Olvi acquired 49.98% of Ragutis shares. Olvi owns Estonian and Latvian breweries Tartu and Cesus. In October 1999, the strategic investor of Vilniaus Tauras became the Danish Brewery Group A/S, and at present owns 95.19% of company shares. In May 2000, the Norwegian company Orkla announced the merge of its brewery Prips Ringnes with Danish Carlsberg, which created the largest brewery in the world Carlsberg Breweries. Orkla group owns 50% of Baltic Beverage Holding shares, which manages the Lithuanian breweries Kalnapilis and Utenos alus. Carlsberg own Klaipėda’s brewery Svyturys. After the merge, Carlsberg Breweries and its companies Svyturys, Utenos alus and Kalnapilis gained the dominant position in the Lithuanian beer market. On 14 September 2001, the Utenos alus, Jungtinis alaus centras and Svyturys companies were reorganised into one joint-stock company Švyturio-Utenos alus. This decision was made during the meeting of shareholders of the companies: Švyturio shareholders own 55.9% of Švyturio-Utenos alus shares, and shareholders of Utenos alus own 44.1%. Baltic Beverages Holding AB sold its 86.6% share of Kalnapilis brewery to Bryggerigruppen A/S (The Danish Brewery Group A/S) at a price of LTL135.1mm. Minority shareholders of whom several are employees, own the remaining share capital. “Kalnapilis was sold due to Lithuanian competition authorities insisting that in order for them to approve the merge of the activities of BBH and Carlsberg in Lithuania, the parties had to sell one of their three breweries. The ownership of BBH is divided 50/50 between Carlsberg Breweries A/S and Oyj Hartwall Abp. On October 6, 2001 Danish brewery The Danish Brewery Group and the then owner of Kalnapilis, - the BBH company, signed the 91

Laimona Sliburyte • Regina Virvilaite contract for the Kalnapilis shares purchase. The competition authorities permitted the Danish Brewery Group to purchase up to 100% of the Kalnapilis’ shares. The Danish Brewery Group currently owns 97.72% of the shares of another Lithuanian brewery Vilniaus Tauras. One may expect even stronger competition within Lithuanian beer companies as further investment in new technologies will take place and new products will be introduced. Also, marketing expenditure will increase. The decision by Scandinavian-owned Baltic Beverages Holding (BBH) to sell part of its assets in Lithuania will change the composition of the beer market in the country. The sale of the Panevezys-based brewery Kalnapilis will surely mean more competition that is likely to drive prices down. The main competitors in the market without any doubt will be Svyturys-Utenos alus and Vilniaus Tauras-Kalnapilis.

Trends in the Lithuanian advertising market In 2002 the real (net) Lithuanian advertising market increased by 5.7% or LTL107m: from LTL187m to LTL197.7m, including mass media and other discount rates. Table 6:2 Analysis of the revenues of media channels for the Lithuanian advertising market in 2000-2001, LTL m. Channel

2000

2001

2001 %

Variation 2000/2001

Television

68

75

38

10.29

Newspapers

66

69

35

4.55

Magazines

23

26

13

13.04

Radio

17

17.7

9

4.12

Outdoor advertising

13

10

5

-23.08

Total

187

197.7

100

5.72

By comparing the Baltic advertising markets, it was calculated that Lithuania lies behind Estonia and Latvia with respect to the amount of money allocated for advertising for one inhabitant. In Lithuania 92

Advertising research: a case study of Lithuanian breweries (3.491million inhabitants) one inhabitant receives Euros 16.3, in Latvia (2.375 million inhabitants) around Euros 26.1, in Estonia (1.372 million inhabitants) Euros 35 are received per inhabitant. The company leaders in advertising in 2001 were: Colgate Palmolive, Procter&Gamble, Coca-cola, Kraft Foods, Bitė, Omnitel, Unilever, Tele 2 and Švyturys. The earnings of these companies made up an even 40% of the total revenues for TV, press, radio and outdoor advertising, calculated in gross prices (without consideration of volume and other discounts, based on official price lists in the mass media). Figure 6:6 Distribution of Lithuanian advertising net revenue according to media channels in 2001

R a d io 9 %

O u td o o r a d v e rtis in g 5 %

T V 3 8 %

M a g a z in e s 1 3 %

N e w sp a p e rs 3 5 %

Calculation based on media channels shows that most of NET revenue last year was received from advertising on TV (38% or LTL75m), newspapers (35% - LTL 69m) and magazines (13% - LTL 26m). Radio and outdoor advertising generated 9% - LTL 17.7m and 5% LTL 10m of revenue. The TV advertising market increased by 10.29%; magazine advertising increased by 13.04%, though outdoor advertising decreased by 23.08%. Advertising in catalogues, on the Internet and in cinemas was not considered.

Purpose of research As our analysis shows, after remarkable changes in the beer market one might expect an even higher competition. Companies find it extremely important to learn about the main factors of competition in the market; whether advertising can play the main role in the competition for consumers; and how the customer perceives advertising.

93

Laimona Sliburyte • Regina Virvilaite This research was designed to answer the following questions: • During the process of differentiation in the Lithuanian beer market, does an interrelationship exist between the company’s expenditure for advertising and its market share? • If the relationship does exist, then how strong is it? • How do consumers perceive advertising of beer? What brands do consumers like most and what is the basis for their choice?

Methodology Quantitative and qualitative research methods were used for the research. To conduct correlation and estimate whether relationship exists between expenditure for advertising and owned market share, Pearson, Kendall’s tau_b, Speraman’s rho correlation coefficients were calculated using SPSS (Statistical Package for Social Sciences, 8; Package of analysis of numbering characteristics) statistical programme. In order to conduct a qualitative research, to research the concept, and branding of advertising, 682 consumers were interviewed. The main requirement for the sample was their actual consumption of beer. A special questionnaire consisting of open-ended and closed questions was prepared.

Findings As our preceding analysis shows, intense competition takes place in the Lithuanian beer market. One of the methods used by competitors in an oligopoly market is the differentiation of products by the use of means of advertising. The following will analyse resources allocated for advertising in the Lithuanian beer industry in 1998-2001, as illustrated in Figure 6:7. Our analysis shows that the Lithuanian beer market expenditure for advertising in 1998 (LTL6.7m) constituted 1.76% of total expenditure (LTL 384m) in Lithuania, and in 1999, about 2.05% (Lithuanian beer industry was LTL10.1m, total expenditure for advertising in Lithuania was LTL 496m). Comparing data for 1998 with data for 1999, it is evident that expenditure for beer advertising increased by a factor of 1.5. In 2000, expenditure continued to increase and reached LTL16.9m, which represented 9.1% of the total expenditure for adver94

Advertising research: a case study of Lithuanian breweries Figure 6:7 Resources allocated for advertising in the Lithuanian beer market in 1998 –2001, LTL million.

16,9

Advertising, LTL mill.

20.000 15.000

17,3

10,1 6,7

10.000 5.000 0

1998

1999

2000

2001

tising in Lithuania. Expenditure for advertising of the companies analysed came to LTL14.2m, which represents an even 84% of the total expenditure for beer advertising in Lithuania. In 2001, a small but still growing increase of expenditure of LTL361.200 was observed for advertising and reached LTL17.3m. Comparing data for the year 2001 with data for the year 2000, there was an increase of 8.8% of Lithuanian expenditure for advertising. The amount allocated by the major Lithuanian beer producers for advertising is illustrated in Figure 6:8. Figure 6:8 Expenditure of the major Lithuanian beer producers for advertising in 1998-2001, LTL Million.

Advertising, LTL mill.

6

5,4

5 4 3

4,6

4,5

4,9

4,4

3,4

2

2,9

3

2,6

2,0

2,1 1,3

1

0,5

0,5

0,2 0,2

0 1998

K alnapilis

1999

U tenos alus

2000

Švyturys

2001

R agutis

95

Laimona Sliburyte • Regina Virvilaite According to the data presented in Figure 6:8, financial allocations for advertising tended to grow. In 1999-2001, Utenos alus allocated the largest amount for advertising. In the same period, Kalnapilis allocated about LTL3m. Švyturys has significantly increased expenditure for advertising in 2001, compared with 1999, from LTL1.3m to LTL4.9m. Ragutis changed its advertising strategy as well: allocations in 1999 were only LTL0.5m, in 2000 they were LTL 2.1m. Figure 6:9 illustrates the comparative weight of expenditure of the main Lithuanian beer producers allocated for advertising within general advertising expenditure of the beer industry. Figure 6:9 Comparative weight of expenditure of the main Lithuanian beer producers allocated for advertising within general beer advertising expenditures in 1998-2001 in percent

Comparative weight of expenditure , %

50

50

1998

1999

2000

2001

44

45 40 35 30 25 20

30

26

31 27

26

28 21

18 17

17

15

13

13 12

12

10 5

5

4

3

Švyturys

Ragutis

3

0

Kalnapilis

Utenos alus

Others

According to the data analysis, three major companies - Utenos alus, Kalnapilis and Švyturys - allocate the largest amount for advertising, which represents respectively from 28% to 50% of the comparative size. Nevertheless, it is evident that allocations are made following the chosen strategy, i.e. they can be smaller and larger if one compares different periods. As our research shows, expenditure for advertising is one of the key actions of competitors in the oligopoly market, therefore we think that it is expedient to conduct a correlative analysis between two variables: expenditure for advertising and owned market share.

96

Advertising research: a case study of Lithuanian breweries

Analysis of Quantitative Data: Correlation Analysis of Expenditure for Advertising and Owned Market Share Data from monthly periods between 1998 and 2001 of the four main Lithuanian companies in the beer market - Kalnapilis, Utenos alus, Švyturys, and Ragutis – were used for the analysis. By marking the market share with RISA98 and further marking every month, the data selection was created; expenditure for advertising was marked as SA98 and further marked every month. By applying the data in calculations, the selection was transferred to the statistics programme SPSS (Statistical Package for Social Sciences, 8; Package of Analysis of Numbering Characteristics). The first thing undertaken was running the Kolmogorov-Smirnov test. Thus the hypothesis H(0), that distribution is normal was tested. Test results showed that this hypothesis cannot be rejected, as the level of significance α was higher than 0.05. As distribution is normal, the Pearson coefficient of correlation was applied for calculations. Pearson’s, Spearman’s and Kendall’s ranks correlation were used to calculate the correlation coefficients. As the results of the calculations take up sixty-three pages, only the cumulative tables are presented in the appendix in order to avoid overloading this chapter. During the correlation analysis, numbering characteristics of the market shares and expenditure for advertising by Utenos alus, Kalnapilis, Švyturys, and Ragutis from 1998 to 2001 were calculated. The market share was marked as RIDAL, and expenditure for advertising was marked as ISL. The results of the analysis on number characteristics of the data shows that in 1998 mean expenditure for advertising totalled LTL1,481,154.90 in 1999, LTL2,248,855.80 in 2000, LTL3,560,610.00 and in 2001, LTL3,422,376.00, and the market share was accordingly 16.275%, 16.675%, 17.55% and 18.2%. The calculated median for number characteristics shows that 50% of all companies in the market do not allocate more resources for advertising (or owned market share) than the suggested median. Calculation of mode allowed definition of the most frequent value of advertising expenditure (owned market share). By calculating the standard deviation, the value of advertising expenditure (owned market share) furthest from the average was defined. During the correlation analysis, 97

Laimona Sliburyte • Regina Virvilaite both minimum and maximum values of analysed data were calculated. The calculation of percentiles has indicated the amount of expenditure for advertising (owned market share), namely 25%, 50%, and 75% of the companies analysed in the beer market. Table 6:3 Pearson, Kendall’s tau_b, Spearman’s rho Correlation Coefficients, 1998 – 2001. Correlation Coefficient

1998

1999

2000

2001

Pearson

0.626

0.661

0.885

0.812

Kendall’s tau_b

0.667

1.000

0.667

0.667

Spearman’s rho

0.800

1.000

0.800

0.800

Results of our research calculations showed that a correlation relationship of medium strong exists between advertising expenditures and market share in 1998 and 1999. As calculated, Pearson’s correlation coefficients fall between ranges of 0.5-0.7. Strong correlation relationship was determined in 2000 and 2001, as the coefficients are higher than 0.7. Also, the correlation coefficients of Spearman and Kendall were calculated. The results of the correlation analysis conducted supports the conclusion that according to the Spearman coefficient, in 1998, 2000, and 2001 a strong correlation relationship existed, and in 1999, the relationship was very strong – perfect positive correlation, as the correlation coefficient was equal to 1. According to the Kendall correlation coefficient, in 1998, 2000 and 2001 the correlation relationship was medium strong, and in 1999, the correlation relationship between expenditure for advertising and owned market share was very strong. This correlation analysis shows that there is a correlation relationship between advertising expenditure and owned market share. As the expenditure for advertising increases, the market share increases as well and vice versa. This provides a basis for the statement that the breweries who are competing in the oligopoly in the Lithuanian beer market and who are allocating more resources for advertising than their competitors can increase the owned market share. The results of the research once again confirm that in an oligopoly, market dominates a non-price competition. 98

Advertising research: a case study of Lithuanian breweries

Analysis of Qualitative Data: The Concept and Branding Exploration The next question the research deals with is how the consumer perceives beer advertising. It is also very important to estimate what brands consumers prefer most and what the reason for their choice is. When analysing the consumers’ perception of beer advertising, it is important to pay attention to one of the aspects of beer consumption habits, i.e. the place or venue where the consumers like to drink beer. Only after assessment of this factor, can reasoned conclusions be made about the formation of specific product concentration towards consumer. In this respect, we are not only interested in the general analysis of the respondents’ answers, but also in analysis according to age and income of the consumers. Figure 6:10 clearly illustrates the differentiation of habits of consumers of different ages. Figure 6:10 Venues for beer consumption according to respondents’ ages. 33

50-59

27

Age

40-49

27

25

30-39 20-29

8

25

17

18

25

28

25

13

13

38

50

18-19 0%

50

10%

20%

Alone at home In bars

30%

8 4

12

4

50 40%

50%

60%

At home with friends Attending public events

70%

80%

90%

100%

On a visit Elsewhere

The graph analysis reflects the tendency that as the age of the group decreases, the venue for beer consumption changes as well. People of fifty most often drink beer at home or when on a visit, whereas among people of forty a tendency to drink beer in bars appears. Venues for beer consumption for people in their thirties includes attending public events and other places, though this group still represents only 20% of the total of beer consumption by this group. In the group of people of 20 to 29 years, consumption of beer in “outdoor” and “indoor”

99

Laimona Sliburyte • Regina Virvilaite environment (bars, public events) changes rather significantly: in consumption of beer by these respondents “outdoors” exceeds 50%. Data for the youngest group is not objective, as the number of respondents for this group is not sufficient. Now we can return to the statement that beer consumed in an “outdoor” environment is draught, while in an “indoor” environment is bottled. This brings us to the conclusion that by targeting consumers of 20-40 years of age, it is essential to develop and implement promotion of beer sold in both shops and bars. Nevertheless, based on sales statistics, the absolute sales of draught beer are still smaller than of the bottled variety; therefore, promotion of bottled beer should remain a priority. Figure 6:11 Distribution of beer consumption according to income

LTL Income

2001+

15

25 29

1501-2000 1001-1500 501-1000

17

Alone at home

18

33 31 20%

24

0

42

19 30%

At home with friends

40%

10

35

23

10%

30

29

20

-500 0%

15

On a visit

60%

70%

In bars

12 8

38 50%

7

6 80%

90% 100%

Attending public events

Analysis of beer consumption venue according to income of consumers also indicated certain tendencies. As income decreases, consumption of draught beer (beer consumption in bars and at public events) decreases. Nevertheless, this tendency, which was clearly observed in the lower income group, changes substantially within the highest income group: here again almost a half of the total beer consumed is draught beer. This supports the conclusion that highestincome consumers visit bars more often than public events, as they have the wherewithal to do so. However, as the largest group of consumers consists of middle-income persons (LTL1001-1500 monthly income), the habits of this group strongly affect the results of the survey. It reflects the main tendency to consume about 65% of bottled beer in an “indoor” environment.

100

Advertising research: a case study of Lithuanian breweries In the case of targeting consumers with a middle income (this group of consumers is quite large and their purchasing power is rather strong), attention to promotion of bottled beer becomes very important. Figure 6:12 Customers’ priorities when selecting a beer

Possibility of buying easily 4% Producer 12%

Product design Advertising 3% Price 2% 17%

Quality 15%

Strength 10% Taste 37%

Among the criteria of analysis which helps to define the reasons for choice of beer, one of the main criteria is the taste of beer. Respondents indicated 37% to importance for this criterion. Price and quality of beer were significantly less important for consumers. It means that when choosing a beer, reputation and the good name of the producer is not that important; the consumption qualities of the product has a greater importance. This supports the conclusion that the producer has to pay a lot of attention to the improvement of taste and the qualitative factors of the beer. The second stage of these efforts (only if the qualitative parameters of the beer are sufficient) should focus on the emphasis of these qualities in advertising. The subconscious of the consumer should contain a positive image of a beer, the qualities of which are associated with a good, reliable taste and high quality. And if these characteristics transferred by advertisement prove themselves when the buyer consumes the beer, then the consumer becomes a permanent buyer of this product. In order to understand the psychology of consumers better, analysis based on respondents’ sex was conducted. The results of this analysis show that men give greater importance to the taste of beer, while women give far less importance to that. The priorities of the women 101

Laimona Sliburyte • Regina Virvilaite interviewed have a more even distribution than men’s priorities, and compared with men, the quality of beer has a critical preponderance (women give an even 20% of importance to quality). For women the price is not as important, although this can be due to the fact that women consume far less beer than men. Here again the different psychologies of the two sexes could be identified: women are more inclined to analyse factors, therefore they understand the taste of beer as a secondary expression of quality and pay more attention to the primary factor – the quality. On the other hand, men are more inclined to make decisions on a practical basis, therefore they perceive the taste of beer as its main characteristic. The fact that the producer of the beer has a greater importance for women than for men, can also serve as a proof that women are more inclined to trust the company image and advertising efforts. An analysis to determine how consumers perceive the matching of a brand of beer with a certain slogan was conducted. Table 6:4 Most suitable slogan to describe a beer.

Carlsberg

Faxe Premium

Gubernija

Kalnapilis

Ragutis

Švyturys

Utenos alus

Vilniaus Tauras

Brands of beer

Seeking for the best

5

0.4

2.7

30.1

1.6

32.5

25.4

2.2

When did you last time…?

1.8

0.7

2.0

10.9

2.7

13.3

66.6

2.1

How are you doing?

5.8

2.7

8.4

19.7

8.3

24.9

25.7

4.6

Probably the best beer in the world?

19.4

1.4

2.2

12.2

1.9

40.2

19.9

2.7

Basketball country’s beer

0.3

0

1.9

13.1

1.9

68.3

13.5

1.0

Legendary men’s beer

1.1

0.6

7.2

7.6

3.1

19.7

12.9

47.8

So natural to be together

1.4

1.0

3.7

14.7

4.1

48.4

23.7

3.0

Let beer talks

2.9

1.7

10.5

13.8

25.8

17.8

18.7

8.7

Always feel certain

5.5

3.9

10.0

17.7

9.3

24.7

23.3

5.5

Danish quality

35.3

45.3

1.0

3.8

1.6

5.4

6.1

1.5

Slogan

We can state that beer advertising slogans used by Utenos, Švyturys, “Vilniaus Tauras, and Ragutis were remembered best by consumers. Nevertheless, some of the slogans used for beer advertising earlier were confused: “Seeking for the best” was used in Kalnapilis advertising, while consumers attribute it to Švyturys – 32.5%. 102

Advertising research: a case study of Lithuanian breweries “Probably the best beer in the world” consumers attribute to Švyturys (40.2%) although Carlsberg used it (19.4%). “Always feel certain” was used in advertising Utenos alus beer, but many consumers (24.7%) attributed the slogan to Švyturys. This analysis shows that a clear leader is Švyturys while other beer brands, even if a little behind, are perceived by consumers differently than stressed in advertising, and this is a rather dangerous situation as some of these consumers can be lost. As the main tool for advertising, beer producers use TV advertising, allocating for it from 60% to 80% of all expenditure for advertising. Respondents were asked what commercials they saw. The data obtained is presented in Table 6:5. Table 6:5 Recollection of beer commercials on TV Advertisement Carlsberg fountain Carlsberg (The beer is tipping into the mug, jottings are changing on the screen) Švyturys ŠNARas Švyturys – the basketball country’s beer Utena When did you last time..?

Have seen the advertisement Yes No 17.2 23.5

82.8 76.5

41.5 63.2 58.6

58.5 36.8 41.4

Most consumers saw the Švyturys beer commercial, and the least number of consumers saw the Carlsberg commercial. Though taking into consideration the fact that the Carlsberg commercial was not run as often as of the Švyturys one, we can make an assumption that this implies less notice being taken of the advertising ffor this brand of beer. An analysis to estimate how the consumers liked beer commercials seen on TV was conducted. Most respondents (57.6%) liked the commercial “When did you last time..?” by Utenos alus where well-known and popular Lithuanians were participating. Although Švyturys was mentioned as the leader in the beer market, in the category of most disliked commercials its commercial “Švyturys ŠNARas” was mentioned most often (19.7%). The latest commercial “Pure water” by Utenos alus unfortunately did not account for much popularity among consumers. The Carlsberg commercial where “beer is being poured into glass, messages change on the screen” was difficult for consumers to evaluate (49.6%). This 103

Laimona Sliburyte • Regina Virvilaite Table 6:6 Evaluation of beer advertising on TV. How did you like the advertisement Advertisement

I didn’t like it

Neither/ nor

Carlsberg fountain

17.6

44.5

I liked it very much 37.9

Carlsberg (The beer is tipping into the mug, jottings are changing on the screen)

17.0

49.6

33.4

Švyturys ŠNARas

19.7

42.9

37.4

Švyturys – the basketball country’s beer

10.4

35.4

54.2

Utena When did you last time..?

10.3

32.1

57.6

analysis shows that advertising ideas were not always successfully applied in the market. In the future, it is important to avoid similar mistakes, as consumers, who are more strongly affected by the competitors’ advertising, can turn to the competitiors’ product. Our analysis of beer brands shows that there are three dominant brands in the beer market at the moment: Švyturys, Utenos and Kalnapilis. They have a similar number of consumers; only Švyturys stands out as having a larger number of loyal consumers. The strong attraction of these brands and loyalty for them shows that consumers are satisfied with the beer these companies offer. Products and images are strong, but there is a threat of reaching a stage where consumption of these brands starts to drop. Although these three brands are in a favourable market situation, their growth potential is not high. It is very important for these three brands to maintain their current consumers and put a lot of effort into paying attention to the needs of the consumers not yet attached to one particular brand. Carlsberg is another brand in the Lithuanian beer market which has similar attractiveness and holds the loyalty of its consumers, although it has a smaller number of consumers. Insufficient possibilities of buying it and not enough familiarity among potential consumers impede the potential growth of Carlsberg. In the future it will be necessary to increase the familiarity of Carlsberg if it wishes to increase its consumers. An analysis of unattached Švyturys consumers allows us to identify a number of their main motives: the best beer produced in Lithuania, tasty beer and highest-quality beer – all these criteria are met by 104

Advertising research: a case study of Lithuanian breweries Švyturys, as are many other criteria. Utenos alus and Kalnapilis are major competitors of Švyturys. Loyal consumers of Švyturys do not consider it as an international beer, while Carlsberg takes a leading role in this respect. They also do not consider Švyturys as being suitable in a situation where there is no better beer; then they drink Ragutis beer. Based on the results of the analysis, the most important reasons for being unattached consumers of Švyturys are: my company beer, tasty and the best Lithuanian beer. Image, campaign priorities and brand position in the market are not important for this group. Švyturys fails within such qualities as: gives peace of mind, produced from the best water, international beer. The main competitor within this segment is Utenos beer. Situations suitable for drinking beer: in the evening, while watching TV, after stress, after work, to have a peaceful rest, to relax and as a tasty drink, to enjoy. Loyal consumers of Švyturys consider this beer as a perfect match to all these situations. In understanding the unattached consumers, the survey reveals that the Švyturys brand fails against other brands on the following qualities: “suitable to suppress thirst” and “suitable for youth”. It has a weak match with “suitable for drinking with friends”; therefore this perception has to be altered, as it is a very important part of image. Consumers associate Švyturys beer with “the basketball country’s beer statement. Research results show that Švyturys risks losing about 3.6% of the market to Carlsberg, while they can attract only about 0.25% consumers from Carlsberg. Currently Švyturys has threats from Kalnapilis and Utenos. Carlsberg and Faxe Premium can cause the main threat in the future. According to the data from our research, the Utenos brand risks losing some of its consumers to the Carlsberg, “Faxe Premium, Kalnapilis, and Švyturys brands. The main reasons for being loyal consumers are: my company beer, tasty beer, beer I would recommend to my friend. Loyal consumers perceive Utenos beer as a perfect match with these criteria. Rather unusually Utenos beer is considered as “having international recognition” or “international beer”. A clear competitor is Švyturys beer, which is closest to Utenos according to most of the qualities analysed. The main reasons for unattached Utenos 105

Laimona Sliburyte • Regina Virvilaite consumers are: tasty beer, my company beer, light to drink, and this beer is not perceived as a good match to these criteria. Both loyal and unattached Utenos consumers identify three main beer consumption situations: a) to relax after stress and work; b) while communicating with friends; c) to suppress thirst. Loyal consumers perceive Utenos beer as most suitable “when one wants to relax”. Utenos beer is not perceived as “beer suitable to drink while watching a basketball match” and “to offer to business partners”. Utenos beer is most often associated with the slogan “When did you last time...?”, and most rarely with basketball. Based on results acquired, Kalnapilis risks losing some of its consumers, who are ready to start consuming Carlsberg and Faxe Premium. Kalnapilis can attract most consumers from Utenos, also from Švyturys. Kalnapilis beer is being perceived as complying with qualities most important for loyal consumers: my company beer, tasty, giving pride. Kalnapilis is not considered an international beer or beer produced from the best water. Kalnapilis is strongly associated with the slogan “In a search for the best!” Its main competitors are the Utenos and Švyturys brands. Analysis supports the conclusion that Carlsberg has the potential to attract a part of the market share from other brands, mostly from Kalnapilis, Švyturys, and Utenos. In this brand, consumers are looking for quality. Loyal Carlsberg consumers like both the Švyturys and the Faxe Premium brands. Unattached consumers of Carlsberg note that this beer is of high quality, tasty and international, although this is not what they want. Carlsberg is considered as a beer for special occasions. Consumers of this brand most often associate it with the slogan “Probably the best beer in the world”. In the future advertising of this brand should lay stress upon requirements for taste, pride and quality.

Conclusions Our research and the results acquired support the following conclusions. We can state that the Lithuanian beer market, after experiencing a growth in 1999, will continue to grow this year as well, only at a slower pace. In the unanimous opinion of the market participants, this 106

Advertising research: a case study of Lithuanian breweries branch of industry – one of a few sectors of industry –will remain as one of the most viable branches of industry in 2002. We can forecast that the branch will continue to grow with tens of millions of Litas investment. The Lithuanian beer market leaders should remain the same, unless Švyturys gets ahead of its competitors Utenos and/or Kalnapilis. Brewery Švyturys has a very strong position in the western part of Lithuania, and if it starts to expand more successfully to other regions, it will acquire a really large growth potential. Change of shareholders of Vilniaus Tauras and Ragutis should positively affect beer quality, marketing strategy and management of these breweries; these companies should maintain similar proportions within the beer market as they do at the moment. Šiauliai Gubernija will maintain the same market share (about 6-9%), although it has implemented large investment projects. Smaller breweries, such as Ragutis and Vilniaus Tauras, will grow at the expense of smaller breweries, only if they succeed in advertising and decreased prices. Hard times are ahead for small Lithuanian breweries – these companies will remain if at all only by having their own niche in provincial areas. Based on our analysis we can state that the Lithuanian beer market has a tight oligopoly structure with its specific characteristics: the three largest breweries own more than 60% of the market. Correlation analysis of the relationship between market share and advertising expenditure in 1998-2001 of Utenos alus, Kalnapilis, Švyturys, and Ragutis was determined. The results generated support the statement that a correlation relationship between advertising expenditure and market share does exist. As the Lithuanian beer market’s structure is a tight oligopoly, the results of correlation analysis obtained prove that non-price means of competition prevail in the oligopoly market. One of such means is differentiation by advertising, which in fact dominates in the Lithuanian beer market. The analysis conducted supports the conclusion that by targeting consumers of the age group between 20-40, it is important to largely develop and implement the promotion of beer sold in shops and bars. Nevertheless, sales statistics show that absolute sales of draught beer are lower than those of bottled varieties; therefore advertising priority should be given to the promotion of the consumption of bottled beer.

107

Laimona Sliburyte • Regina Virvilaite When targeting middle-income consumers (this group is rather large and its purchasing power is strong), we find it important to pay more attention to bottled beer advertising. Beer which is associated with the words: good, reliable taste and high quality, must form a positive image in the consumers’ subconscious. And if these characteristics, stated in the advertising, prove themselves to be true while the buyer consumes the beer, then that person becomes a loyal consumer of this product. Analysis of beer brands shows that in the beer market currently three brands dominate: Švyturys, Utenos and Kalnapilis. They have similar numbers of consumers; only Švyturys stands out as having a larger number of loyal consumers. The high attractiveness of these brands and the attachment to them shows consumers’ satisfaction with the beer. Brands and imagines are strong, yet there is a risk of reaching a stage where consumption of these brands starts to drop. Although these three brands are in a favourable market situation, yet potential of growth is not high. It is important for these brands to maintain current consumers and pay more attention to the needs of not yet attached consumers. Analysis of attractiveness and loyalty to beer brands shows the strong market positions of Švyturys, Kalnapilis, Utenos and Carlsberg. It is evident that in this respect Švyturys is in the strongest position. Yet, all these brands, except Carlsberg do not have much potential for growth. Other beer brands - Ragutis, Gubernija and Vilniaus Tauras have lower than medium loyalty and this is caused by problems of the product. This should receive the primary and the greatest attention. Three market leaders – Utenos, Švyturys and Kalnapilis – share the largest market share. In order to avoid loss of market shares, these brands should focus more on current consumers instead of nonconsumers. These brands have a large segment of loyal consumers, their products are strong and therefore it is recommended that they increase mass advertising. Such strengths of brands as quality, taste, patriotism (pride), and socialisation should be emphasised and presented in advertising for target groups. Carlsberg beer is related with a different market segment. This beer is associated with quality, taste, internationalisation, and elitism. This brand still has a considerably small number of consumers; therefore, it

108

Advertising research: a case study of Lithuanian breweries is important to increase the familiarity of the brand by increasing advertising and by stressing its positive qualities.

Appendix Correlations RIDAL98 RIDAL99 RIDAL00 RIDAL01 RIDAL98 Pearson Correlati 1,000 ,675 ,586 ,557 Sig. (2-tailed) , ,325 ,414 ,443 N 4 4 4 4 RIDAL99 Pearson Correlati ,675 1,000 ,888 ,759 Sig. (2-tailed) ,325 , ,112 ,241 N 4 4 4 4 RIDAL00 Pearson Correlati ,586 ,888 1,000 ,971* Sig. (2-tailed) ,414 ,112 , ,029 N 4 4 4 4 RIDAL01 Pearson Correlati ,557 ,759 ,971* 1,000 Sig. (2-tailed) ,443 ,241 ,029 , N 4 4 4 4 ISL98 Pearson Correlati ,626 ,583 ,169 -,023 Sig. (2-tailed) ,374 ,417 ,831 ,977 N 4 4 4 4 ISL99 Pearson Correlati -,009 ,661 ,403 ,188 Sig. (2-tailed) ,991 ,339 ,597 ,812 N 4 4 4 4 ISL00 Pearson Correlati ,190 ,804 ,885 ,810 Sig. (2-tailed) ,810 ,196 ,115 ,190 N 4 4 4 4 ISL01 Pearson Correlati ,311 ,886 ,910 ,812 Sig. (2-tailed) ,689 ,114 ,090 ,188 N 4 4 4 4

ISL98 ,626 ,374 4 ,583 ,417 4 ,169 ,831 4 -,023 ,977 4 1,000 , 4 ,511 ,489 4 ,054 ,946 4 ,206 ,794 4

ISL99 -,009 ,991 4 ,661 ,339 4 ,403 ,597 4 ,188 ,812 4 ,511 ,489 4 1,000 , 4 ,676 ,324 4 ,718 ,282 4

ISL00 ,190 ,810 4 ,804 ,196 4 ,885 ,115 4 ,810 ,190 4 ,054 ,946 4 ,676 ,324 4 1,000 , 4 ,987* ,013 4

ISL01 ,311 ,689 4 ,886 ,114 4 ,910 ,090 4 ,812 ,188 4 ,206 ,794 4 ,718 ,282 4 ,987* ,013 4 1,000 , 4

*. Correlation is significant at the 0.05 level (2-tailed).

109

Laimona Sliburyte • Regina Virvilaite Correlations Kendall's tau_ RIDAL98 Correlation Coeffic Sig. (2-tailed) N RIDAL99 Correlation Coeffic Sig. (2-tailed) N RIDAL00 Correlation Coeffic Sig. (2-tailed) N RIDAL01 Correlation Coeffic Sig. (2-tailed) N ISL98 Correlation Coeffic Sig. (2-tailed) N ISL99 Correlation Coeffic Sig. (2-tailed) N ISL00 Correlation Coeffic Sig. (2-tailed) N ISL01 Correlation Coeffic Sig. (2-tailed) N Spearman's rh RIDAL98 Correlation Coeffic Sig. (2-tailed) N RIDAL99 Correlation Coeffic Sig. (2-tailed) N RIDAL00 Correlation Coeffic Sig. (2-tailed) N RIDAL01 Correlation Coeffic Sig. (2-tailed) N ISL98 Correlation Coeffic Sig. (2-tailed) N ISL99 Correlation Coeffic Sig. (2-tailed) N ISL00 Correlation Coeffic Sig. (2-tailed) N ISL01 Correlation Coeffic Sig. (2-tailed) N

RIDAL98 RIDAL99 RIDAL00 RIDAL01 ISL98 1,000 ,333 ,333 ,333 ,667 , ,497 ,497 ,497 ,174 4 4 4 4 4 ,333 1,000 ,333 ,333 ,667 ,497 , ,497 ,497 ,174 4 4 4 4 4 ,333 ,333 1,000 1,000* ,000 ,497 ,497 , ,042 1,000 4 4 4 4 4 ,333 ,333 1,000* 1,000 ,000 ,497 ,497 ,042 , 1,000 4 4 4 4 4 ,667 ,667 ,000 ,000 1,000 ,174 ,174 1,000 1,000 , 4 4 4 4 4 ,333 1,000* ,333 ,333 ,667 ,497 ,042 ,497 ,497 ,174 4 4 4 4 4 ,000 ,667 ,667 ,667 ,333 1,000 ,174 ,174 ,174 ,497 4 4 4 4 4 ,000 ,667 ,667 ,667 ,333 1,000 ,174 ,174 ,174 ,497 4 4 4 4 4 1,000 ,400 ,400 ,400 ,800 , ,600 ,600 ,600 ,200 4 4 4 4 4 ,400 1,000 ,400 ,400 ,800 ,600 , ,600 ,600 ,200 4 4 4 4 4 ,400 ,400 1,000 1,000** ,200 ,600 ,600 , , ,800 4 4 4 4 4 ,400 ,400 1,000** 1,000 ,200 ,600 ,600 , , ,800 4 4 4 4 4 ,800 ,800 ,200 ,200 1,000 ,200 ,200 ,800 ,800 , 4 4 4 4 4 ,400 1,000** ,400 ,400 ,800 ,600 , ,600 ,600 ,200 4 4 4 4 4 ,200 ,800 ,800 ,800 ,400 ,800 ,200 ,200 ,200 ,600 4 4 4 4 4 ,200 ,800 ,800 ,800 ,400 ,800 ,200 ,200 ,200 ,600 4 4 4 4 4

*. Correlation is significant at the .05 level (2-tailed). **.Correlation is significant at the .01 level (2-tailed).

110

ISL99 ISL00 ISL01 ,333 ,000 ,000 ,497 1,000 1,000 4 4 4 1,000* ,667 ,667 ,042 ,174 ,174 4 4 4 ,333 ,667 ,667 ,497 ,174 ,174 4 4 4 ,333 ,667 ,667 ,497 ,174 ,174 4 4 4 ,667 ,333 ,333 ,174 ,497 ,497 4 4 4 1,000 ,667 ,667 , ,174 ,174 4 4 4 ,667 1,000 1,000* ,174 , ,042 4 4 4 ,667 1,000* 1,000 ,174 ,042 , 4 4 4 ,400 ,200 ,200 ,600 ,800 ,800 4 4 4 1,000** ,800 ,800 , ,200 ,200 4 4 4 ,400 ,800 ,800 ,600 ,200 ,200 4 4 4 ,400 ,800 ,800 ,600 ,200 ,200 4 4 4 ,800 ,400 ,400 ,200 ,600 ,600 4 4 4 1,000 ,800 ,800 , ,200 ,200 4 4 4 ,800 1,000 1,000** ,200 , , 4 4 4 ,800 1,000** 1,000 ,200 , , 4 4 4

7

Advertising and the Prominence of the Corporate Brand Identifying Brand Architecture Use through Content Analysis KIM CRAMER, PETER C. NEIJENS & EDITH G. SMIT

The corporate brand has become increasingly important in today’s markets and society. Not only do corporate brands provide companies with competitive advantages, they also represent the way in which companies behave towards society. Products and brands have been fulfilling consumers’ needs for quite some time, but now, as markets are full of interchangeable products and brands, consumers increasingly look for distinguishing companies. Companies use a corporate branding strategy - that is, a strategy in which the corporate brand plays a central role - to radiate reliability, innovativeness, or social responsibility, for example. Consumers use such company associations to evaluate the company. This evaluation is often referred to as the corporate image. Consumers’ evaluations of the company’s products and services are generally considered to be moderated by this corporate image and vice versa; the product evaluations influence the image of the corporate brand. This process is called brand value transfer or association transfer (e.g., Brown & Dacin 1997; Keller & Aaker, 1992; Wansink, 1989; Yoon, Guffey & Kijewski, 1993). The influence of the corporate image on the evaluation of a company’s products or services partly depends on the degree with which the corporate brand is visible in the communication of products or services, e.g. advertising. A prominently visible corporate brand increases the brand value transfer effect, because it provides easier access to the corporate associations that are present in consumers’ 111

Kim Cramer • Peter C. Neijens • Edith G. Smit minds (Berens & Van Riel, 2001). This indicates that if companies want consumers to transfer corporate associations to the company's products or services, they should show their corporate brand name in product advertising. However, as Herbig and Milewicz (1995, p. 9) say, using the corporate brand “may be counterproductive if buyers see no meaningful relationship between the old and new business and brand reputation transference is not successful.” The effect that a negative product evaluation has on the corporate brand has been studied under the name dilution (e.g., Ahluwalia & Gürhan-Canli, 2000; Milberg, Whan Park, & McCarthy, 1997; Roedder John, Loken, & Joiner, 1998). To reduce the risk of dilution, companies should try to prevent the transfer of associations by not using the corporate brand in product advertising. In spite of the growing importance of corporate branding, there is not much research explaining the determinants and effects of corporate branding. However, for managers to understand how corporate branding can benefit their companies, it is important to have insight into both the determinants and effects of corporate branding, and in the actual prominence of the corporate brand in the company’s communication. In this study, we have used content analysis to investigate the prominence of corporate brands in the advertisements of companies operating in two service markets: the temporary labor market and the financial services market (banks and insurance companies). By looking at the brand architecture used in each advertisement - that is, the combination of brands representing a product or service -, we learned whether or not the corporate brand was used, and if so, how prominent it was. We also investigated the factors that influence the use of certain brand architectures. To the best of our knowledge, no previous studies have analyzed advertising contents from the perspective of brand architecture. With this study, we aim to contribute to the existing branding literature by investigating corporate brand prominence and discovering determinants of corporate branding. Our central research question is: Which factors determine the prominence of the corporate brand in advertising?

112

Advertising and the Prominence of the Corporate Brand Three questions further develop the central research question: RQa: Which brand architectures are used in advertising? RQb: To what degree is the corporate brand visible in advertising? RQc: Which factors are connected with corporate brand prominence? First, we will present a short overview of the relevant literature in the field of corporate branding and its determinants. Second, we will present our content analysis of brand architectures in advertising. Third, we will look at previous studies in which we used desk research and interviews to uncover the factors that influence the brand portfolio strategies of the selected companies in the temporary labor market and the financial services market (Blom & Cramer, 2002; Cramer, 2001). We will be linking the data of these studies with the data of the content analysis to find determinants that are connected with corporate brand prominence in advertising. Finally, we will discuss our findings.

Corporate Branding There are two viewpoints when looking at corporate branding: the viewpoint of corporate image (e.g., Brown, 1998; Maathuis, 1999; Van Riel, 1994) and the viewpoint of brand portfolio management (e.g., Laforet & Saunders, 1999). In this section, we will first look at the corporate brand from the viewpoint of corporate image, and then the viewpoint of brand portfolio management.

The Role of the Corporate Brand in Corporate Image The first builders of - what we now call - corporate image were architects and designers. By integrating the design of indoor and outdoor furniture, equipment, and trademarks of shops in the late 1800’s, they attempted “to influence people’s perceptions of a company through consistent presentation of the company to its various publics” (Brown, 1998, p. 216). Since the late 1950’s, the notion of the existence of corporate identity and personality has existed in the industry; however, there has been some debate in the marketing literature about the definition of the corporate image concept. Related concepts like corporate reputation, corporate identity, corporate

113

Kim Cramer • Peter C. Neijens • Edith G. Smit personality, and brand image, make it hard to define corporate image, because it is not clear what the differences and similarities between the concepts are (see Brown (1998) and Patterson (1999) for an overview of related concepts and definitions). Brown (1998) observes three key characteristics of corporate image: first, it exists in the minds of individuals and may differ from individual to individual; second, there are different audiences for any company, such as consumers, financial analysts, and shareholders; third, various psychological phenomena are believed to form a corporate image. For example, some authors think of corporate image as a person’s perception of a company (e.g., Carlson, 1963; Gronroos, 1984), or a mental picture (e.g., Britt, 1971; Hardy, 1970). Others include feelings, attitudes, or evaluations of a company in their definition of the concept (Barich & Kotler, 1991; Keller, 1993). To circumvent the definitional confusion, Brown and Dacin (1997) refer to corporate associations as a collective label for all the information individuals have about a company, be it cognitive, affective, or evaluative. The services marketing literature has paid little attention to the concept of corporate image (De Ruyter & Wetzels, 2000). This is surprising because the intangibility of services leads to a greater dependency on company reputation than tangible products that can be evaluated on the basis of look, feel, taste, and smell. According to De Chernatony and Dall’Olmo Riley (1999), consumers use the company’s size and reputation as criteria for service quality. The lack of attention for corporate image and service branding as concepts in the service literature can be explained by the fact that, traditionally, the service industries have not been very brand-oriented (Camp, 1995). For example, consumers do not often regard the names of financial companies as (corporate) brand names (Kapferer, 1995). This is illustrated by the fact that no financial company is listed in the top ten, or even twenty, of the world’s most valuable brands (Day, 1997). A good reputation is important for companies, especially service companies, to have consumers evaluate the company and its products and services positively. One strategy for reputation building is the use of corporate branding: “the process of creating and maintaining a favorable reputation of the company and its constituent elements, by sending signals to stakeholders using the corporate brand” (Maathuis 114

Advertising and the Prominence of the Corporate Brand 1999, p. 5). ‘Constituent elements’ can e.g. be business units, subsidiaries, products and services of the company. Stakeholders are people or groups that have a relationship with the company, such as consumers, shareholders, suppliers, retailers, trade organizations, journalists, regulators, and organizational members. Harris and De Chernatony (2001) add that, to create this favorable reputation, the messages about the brand identity should be consistent and the delivery uniform across all stakeholder groups. Figure 7:1 Relationship between brand portfolio strategy and corporate brand prominence. importance of corporate brand

mono

endorsement

multi

The Role of the Corporate Brand in Brand Portfolio Management If we look at corporate branding from the second viewpoint, brand portfolio management, we find that the role of the corporate brand varies across different brand portfolio strategies. A brand portfolio consists of a company’s brands, products, and services. To manage a brand portfolio effectively, managers must make, implement, and monitor choices regarding the number of and relationships between brands, products, and services (Den Engelsen & Hurts, 2001). In general, three brand portfolio strategies are distinguished in the literature (Douglas, Graig, & Nijssen, 1997; Kapferer, 1992; Olins, 1989; Van Riel, 1994). In a mono brand strategy, the corporate brand name is used for all products and services of the company. In a multi brand strategy, every product or service has its own individual brand 115

Kim Cramer • Peter C. Neijens • Edith G. Smit name. In an endorsement strategy, a product or service combines its own individual brand name with the corporate brand name. Some authors distinguish several forms of endorsement branding, for example dual branding (Franzen, 2000; Laforet & Saunders, 1999), or weak versus strong endorsement (Berens & Van Riel, 2001). With dual branding, the individual brand and the corporate brand are combined with equal emphasis on both brands. A weak endorsement is characterized by less emphasis on the corporate brand, while a strong endorsement is characterized by more emphasis on the corporate brand. The mono endorsement and multi brand strategies thus form a continuum, with the corporate brand being less important towards the direction of multi branding. Figure 7:1 shows the relationship between brand portfolio strategy and corporate brand prominence. Virgin is a classic example of a company that uses a mono brand strategy. All Virgin products or services, be it airline companies, wedding dresses, or holidays, wear the Virgin brand name, sometimes accompanied by a label (e.g., Virgin Atlantic, Virgin Brides, or Virgin Holidays). Sometimes it is accompanied by a label, like Virgin Brides, but it is always Virgin. To the right side of the continuum, an example of dual branding is Centraal Beheer Achmea (Centraal Beheer is an insurance brand of the Achmea company). An example of endorsement branding is the way the individual Fortis brand MeesPierson is connected to Fortis, by mentioning that MeesPierson is part of the Fortis group. An example of a company that uses a multi brand strategy is Procter & Gamble. Most of their products have their own individual brand name. The corporate Procter & Gamble brand is not prominently visible.

Determinants of Corporate Brand Prominence Several authors have investigated the factors that influence marketing, advertising and branding strategies (Aaker, 1996; Cramer, 2000, 2001; Kapferer, 1992; Keller, 1998; Kotler & Armstrong, 1993; Laforet & Saunders, 1999; Maathuis & Van Riel, 1996; Van Riel & Berens, 2000). Kotler and Armstrong (1994, p. 69) define the marketing surroundings of a firm as “the actors and forces that are not part of marketing itself, but which have an influence on the success of the transactions between the company and its target groups”. Figure 7:2 116

Advertising and the Prominence of the Corporate Brand presents an overview of the factors mentioned in the literature, categorized into three groups: characteristics of the company, objectives of the company, and environment of the company. The objectives of the company can sometimes be determinants and effects at the same time, as indicated by the dashed arrow. For example, a company’s wish to increase employee involvement may be a determinant leading to a mono brand strategy (strong corporate brand prominence), but employee involvement may also be an effect, evolving from a mono brand strategy. In our study, it was not possible or feasible to test all expectations about the factors that determine the corporate brand prominence, as presented in Figure 7:2. For example, we could not test the expectation that companies in homogeneous markets use their corporate brand more prominently than companies in heterogeneous markets, since both the temporary labor market and the financial services market are homogeneous. In this paper, we investigated three factors on the basis of which the companies could be distinguished and which managers of the companies regard as highly important: available budget, degree of fit between products and core business, and globalization. Available budget and degree of fit between products and core business belong to the category ‘characteristics of the company’. Globalization belongs to the category ‘objectives of the company’. Our previous studies (Blom & Cramer, 2002; Cramer, 2001) gave us insight into these independent variables belonging to the selected companies. The three factors will be briefly introduced.

Advertising Budget One of the factors influencing brand portfolio strategy is the available budget for brand development and communication. Building and supporting a brand is extremely expensive; building and supporting more than one brand is even more expensive. Cost efficiency often leads to mono branding, because this brand portfolio strategy reduces the costs of supporting the brand. The investments in production, design, packing, and advertising need to be done for only one brand, allowing the company to promote all of its products, while advertising only one (Murphy, 1987). Therefore, we expect that companies with

117

Kim Cramer • Peter C. Neijens • Edith G. Smit small advertising budgets make more use of corporate branding than companies with large advertising budgets (H1). Figure 7:2. Determinants of corporate brand prominence. Characteristics of the company Long-term ownership (+) History (+) Available brand development and communication budget (-) Product range (-) Degree of fit between products and core business (+) Compatibility of technology (+) Compatibility of markets and target audiences (+)

Environment of the company Degree of market consolidation (+)

Degree of fit between individual brand and corporate brand (+) Degree of centralization (+)

Degree of homogeneity of the market (+) Equity of acquired brand (-) Activities of the competition (+/-) Stakeholder needs (+/-) Degree of corporate brand prominence

Retailer dependency (+/-) Demographics (+/-) Economic developments (+/-) Technological developments (+/-)

Objectives of the company

Politics (+/-)

Economies of scale (+) Visibility in financial market (+) Globalization (+) Attract employees (+) Increase employee involvement (+) Brand association transfer (+) Market segmentation (-) Internal competition (-) Risk reduction (-)

(+) Positive relationship. For example: the higher the degree of centralization, the higher the degree of corporate brand prominence. (-) Negative relationship. For example: the wider the product range, the lower the degree of corporate brand prominence. (+/-) Neutral relationship. For example: if the competition moves towards mono branding, the company can either follow or not; there is no clue about the direction of the influence.

118

Advertising and the Prominence of the Corporate Brand

Degree of Fit between Products and Core Business Fit is one of the most discussed concepts in the literature on brand portfolio management. The concept of fit is defined in many different ways by different authors (e.g., Aaker & Keller, 1990; Franzen & Van den Berg, 2002; Loken & Roeder John, 1993, Sheinin, 1998). In general, it means similarity between two phenomena. Aaker and Keller (1990) distinguish three components of fit: Complement (the extent to which consumers view two product classes as complements), Substitute (the extent to which consumers view two product classes as substitutes) and Transfer (the perceived ability of a firm operating in a product class to make a product in another product class). In contrast to Aaker and Keller, Loken and Roedder John (1993) do not explicitly define fit, but use the concept of typicality of brand extensions to relate to the same concept. When a new product is introduced under an existing brand name (brand extension), consumers may perceive this product as either typical or atypical for the family brand. This interpretation mostly resembles the Transfer component of fit: does the new product or service logically match the company’s core business? For financial services, Blom and Cramer (2002) suggest a continuum that ranges from banking services, via all finance services (both banking and insurance services), to insurance services. Companies can be placed on this continuum according to their core business; they are originally banks or insurance companies. Some companies diverge from their original core business and move towards the all finance position in the middle of the continuum, which reduces fit. When a new product is introduced under its own individual brand name, it may still be desirable to link it to the company by means of endorsement branding. Additionally, a different kind of fit is in question here: the fit between the individual brand and the corporate brand. If the values of both brands do not match, endorsement branding is less used. An example that illustrates the importance of matching brand values is Nutricia, a firm that sells children’s nutrition, medical food supplements, and natural foods. The natural foods have the brand name Zonnatura. Nutricia does not endorse them, because the brand values of Zonnatura – ‘as natural as possible, no additives’ – do not match with those of Nutricia: ‘added vitamins’ (Hoogeweegen, 2000).

119

Kim Cramer • Peter C. Neijens • Edith G. Smit In our study, we focus on the meaning of fit as the degree in which the services provided by the company match its core business. We expect that companies with a range of different kinds of services (low degree of fit between product and core business) make less use of corporate branding than companies with similar services (large degree of fit between products and core business) (H2).

Globalization Due to political and economic integration, a growing global market infrastructure and the growing mobility of consumers and other stakeholders, geographic boundaries are fading (Douglas, Craig, & Nijssen, 2001). This has led companies to increase global activities. If globalization is an objective, companies generally use one brand name to be strongly represented all over the world. Therefore, we expect that companies with global activities make more use of corporate branding than companies that are not globally active (H3).

Study To answer the central research question, which factors determine the prominence of the corporate brand in advertising, we conducted a quantitative content analysis. Here our aim was to answer the first two questions: RQa) which brand architectures are used in advertising? and RQb) to what degree is the corporate brand visible in advertising? In the content analysis, we used companies’ advertising messages as the units of analysis. Second, we connected the data of the content analysis with information about the companies’ budget, degree of fit between products and core business and global activity, which we obtained through desk research and interviews with managers in previous studies (Blom & Cramer, 2002; Cramer, 2001), to answer the third question: RQc) which factors are connected with corporate brand prominence?

Corporate Brand Prominence: Content Analysis Sample Advertisements from eighteen companies that appeared in various media during a one-year period (from July 2000 to August 2001) were 120

Advertising and the Prominence of the Corporate Brand analyzed. The companies were selected within two Dutch markets: the temporary labor market (7 companies) and the financial services market (11 companies). The selection of markets and firms had already been carried out in our previous related studies concerning the determinants of brand portfolio strategies (Blom & Cramer, 2002; Cramer, 2001). We selected service markets because branding research has principally focused on physical goods, leaving the possible problems and challenges of service branding unexplored. Both the temporary labor and the financial services market have easily identifiable leading firms. Recent brand changes in these markets made it relatively easy to identify underlying considerations while interviewing the managers. Only firms that operate and advertise on a national scale were selected; most companies operate worldwide or mainly in Europe. The selected period for the content analysis corresponds with the period during which managers were interviewed about the determinants of their company’s brand portfolio strategy (Blom & Cramer, 2002; Cramer, 2001). The advertisements were gathered with the help of an advertising agency (PPGH/JWT), using a national advertising database and an archive of print ads. The media in which the advertisements appeared were television, radio, newspapers, consumer magazines, trade magazines, and outdoor media. Almost every company advertised more than one brand. The sample we took contained one advertisement per brand per medium. We decided to select the most expensive advertisement that had appeared most frequently, assuming that both costs and frequency indicate the importance of the ad. The sampling procedure led to a sample of 368 ads.

Codebook A codebook containing 33 variables was used to analyze the advertisements (see Appendix A). To determine the brand architectures used (RQa), we looked at the number of brands, the nature of the brands, and the combinations of brands. To specify the number of brands within an advertisement, we simply counted and listed the brands (variable no. 17 in the code book). To define the brands’ nature, the brands were included in one of the categories ‘corporate brand’, ‘individual brand’, ‘sub brand’, or ‘label’1 (variable 121

Kim Cramer • Peter C. Neijens • Edith G. Smit no. 20). To determine how brands were combined, we listed the (parts of) sentences in which two or more brands were mentioned (nos. 29 and 30). Also, coders included the advertisement in one of fifteen brand architecture categories, as identified in Box 1 (no. 33). Kok (1997) presents four categories: 1) corporate brand; 2) corporate brand with support from product brand; 3) product brand with support from corporate brand; 4) product brand. Several other authors also suggest such a typology of possible brand combinations, most of them being quite comparable (Bath, Kelley, and O’Donnell, 1998; Saunders & Guoqun, 1996; Van Riel & Berens, 2000). Franzen (2000) distinguishes seven variants of brand combinations, featuring main brands, endorsement brands, sub brands, and labels. To be able to analyze the advertisements as precisely as possible, we combined and adapted the various categories, resulting in the brand architecture typology presented in Box 1. In the first category, an individual brand is used as a stand-alone brand. This type of brand does not necessarily need or want support from the company behind it. Consequently, with individual branding the corporate brand is not present in the advertisement. An example is ‘Visa Cards’2. A sub brand or a label, or both, may accompany the individual brand. A sub brand is a subordinate brand that is added to a main brand (corporate or individual). It adds meaning to the brand combination (Franzen, 2000), but it is never used alone. An example is ‘Visa Cards FlexInvest’, in which FlexInvest is a sub brand representing a flexible form of investing money. The difference between sub brands and labels is that a label is not a brand. A brand is the property of the company, while a label is ‘everyday language’, which cannot be claimed (Franzen, 2000). It is a generic description that helps to identify a product, e.g. ‘Visa Cards Business’ (category 1c) or ‘Visa Cards FlexInvest Business’ (category 1d). In this case, the addition ‘Business’ is not a brand, but it explains that this product is especially meant for the business user. In the second category, the individual brand is visibly supported (endorsed) by the corporate brand, e.g. ‘Visa Cards is a member of Fortis’ or ‘Visa Cards by Fortis’. The emphasis is on the individual brand. Sub brands and labels may be used to further identify the products or services. In the third category, dual branding, the emphasis is on both the individual and the corporate brand, as is the case with ‘Visa Cards Fortis’ or ‘Fortis Visa 122

Advertising and the Prominence of the Corporate Brand Cards’. It is obvious that in this case a choice must be made as to which brand should be mentioned first and which last. Box 7:1. Brand Architecture Typology 1) Individual branding a) Only individual brand b) Individual brand + sub brand c) Individual brand + label d) Individual brand + sub brand + label 2) Endorsement branding (emphasis on individual brand) e) Corporate brand + individual brand f) Corporate brand + individual brand + sub brand g) Corporate brand + individual brand + label h) Corporate brand + individual brand + sub brand + label 3) Dual branding (emphasis on both brands) i) Corporate brand + individual brand j) Individual brand + corporate brand 4) Corporate branding k) Corporate brand + sub brand l) Corporate brand + label m) Corporate brand + sub brand + label n) Only corporate brand

The emphasis on each brand is thus not exactly the same (the first brand being more emphasized than the last). In the last category, the corporate brand is used, with or without additions. An example of category 4n would simply be: Fortis. The four main brand architecture categories correspond with the three brand portfolio strategies (mono, endorsement, and multi brand strategy). If a company uses a mono brand strategy, we expect that only the corporate brand is used in the company’s advertisements. So, the advertisements are expected to be included in category 4. In case of a multi brand strategy, we expect that the advertisements contain only individual brands (category 1). In case of an endorsement strategy, we expect that a combination of the corporate brand and the individual brand is used. This is the case in both category 2 (endorsement branding) and 3 (dual branding).

123

Kim Cramer • Peter C. Neijens • Edith G. Smit The prominence of the brands, with special attention on the corporate brand (RQb), was measured in two ways: objectively and subjectively. The objective variables were: the frequencies with which the brand(s) were verbally and visually used, the duration of the use (in seconds)3, whether or not a logo was shown, the relative size of the brand(s), the physical position of the brand, and the colors used (no. 21 to 27). If a brand is most often mentioned orshown, and if it is shown the longest, the biggest, in a prominent position, and intensely colored, it can be considered the most important brand of that ad. In addition, in a more direct but also more subjective way, the coders were asked to identify the most prominent brand (no. 31) and to classify the degree of prominence of the corporate brand on a 5-point scale (from ‘not at all prominent’ to ‘very prominent’, no. 32). Other variables in the codebook were, among others, market, company, advertising medium, appearance frequency, and the mentioning of a specific product or service in the ad. Three coders were instructed. After testing the validity and usability of the codebook, each of the coders encoded a part of the sample. To determine inter-coder reliability, 60 ads (16 % of the total number of ads) were encoded by all coders.

Company Information: Desk Research and Interviews It was not possible or feasible to test all expectations about the factors that determine the corporate brand prominence, as presented in Figure 7:2. For example, we could not test the expectation that companies in homogeneous markets use their corporate brand more prominently than companies in heterogeneous markets, since both the temporary labor market and the financial services market are homogeneous. We chose to use three factors based on which the companies could be distinguished and which the company managers regard as highly important: advertising budget, degree of fit between products and core business, and globalization. Previous studies, in which we interviewed the managers of the eighteen companies after extensive desk research (Blom & Cramer, 2002; Cramer, 2001), gave us insight into these variables of the selected companies. First, to test the assumption that globally active companies make more use of corporate branding than companies that are not globally active, we categorized the companies 124

Advertising and the Prominence of the Corporate Brand according to whether they were globally active or not. Companies were categorized as globally active if (most of) their brands, and their corporate brand as a minimum, were represented around the world. Second, we expected companies with large budgets to make more use of individual branding than companies with small budgets. Therefore, we categorized the companies as having either a large advertising budget or a small advertising budget, based on the total advertising expenditures in guilders, in the selected one-year period. Third, we expected that companies with a range of different kinds of services (low degree of fit between product and core business) would make more use of individual branding and less use of corporate branding than companies with comparable services (large degree of fit between products and core business). To categorize the companies, we used the continuum for financial services suggested by Blom and Cramer (2002), which ranges from banking services, via all finance services (both banking and insurance services), to insurance services. Companies providing only banking services or only insurance services were both categorized as offering comparable services (fit). Companies providing both banking and insurance services (all finanz services) were categorized as offering different kinds of services (no fit). In the temporary labor market, only a few companies provide other services than temporary labor, for example education, childcare, and cleaning services. These companies were categorized as having a low degree of fit between products and core business.

Results Inter-coder reliability was established using Cohen’s Kappa4. When Kappa values were low, the coders reconsidered their coding performance together and decided which was the best value for the variables in specific cases. However, after that, some variables still had very low Kappa values (